modulation of gamma oscillatory activity through ... · ii modulation of gamma oscillatory activity...
TRANSCRIPT
Modulation of Gamma Oscillatory Activity Through Repetitive Transcranial Magnetic Stimulation in Healthy
Subjects and Patients with Schizophrenia
by
Mera Sun Barr
A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy
Institute of Medical Science University of Toronto
© Copyright by Mera Sun Barr 2011
ii
Modulation of Gamma Oscillatory Activity Through Repetitive
Transcranial Magnetic Stimulation in Healthy Subjects and
Patients with Schizophrenia
Mera Sun Barr
Doctor of Philosophy
Institute of Medical Science
University of Toronto
2011
Abstract
Background: Gamma oscillations (30-80 Hz) in the dorsolateral prefrontal cortex (DLPFC) are
associated with working memory; a process involving the maintenance and manipulation of
information on line (Baddeley, 1986). Gamma oscillations are supported by gamma-
aminobutyric acid (GABA) inhibitory interneurons in the DLPFC. Repetitive transcranial
magnetic stimulation (rTMS) is a non-invasive method in which to stimulate the cortex that has
been shown to modify oscillations, cognition and GABAergic mechanisms. Patients with
schizophrenia have severe deficits in working memory that may be related to impairments in
GABAergic inhibitory neurotransmission underlying gamma oscillations in the DLPFC.
Objective: First, to evaluate gamma oscillatory activity in patients with schizophrenia during
working memory compared to healthy subjects. Second, to examine the effect of rTMS applied
over the DLPFC on gamma oscillations generated during working memory in healthy subjects.
Third, to examine the effect of rTMS applied to the DLPFC on gamma oscillations in patients
with schizophrenia compared to healthy subjects. Hypotheses: First, it was hypothesized that
patients with schizophrenia would exhibit an alteration in gamma oscillatory activity. Second, it
was hypothesized that rTMS would be effective in enhancing gamma oscillations in healthy
iii
subjects. Third, it was hypothesized that rTMS would be effective in inhibiting gamma
oscillations in patients with schizophrenia. Results: The first study found that patients with
schizophrenia generate excessive gamma oscillations during working memory compared to
healthy subjects. The second experiment found that rTMS over the DLPFC resulted in the
potentiation of gamma oscillations in healthy subjects during working memory. The third
experiment demonstrated that rTMS inhibited excessive gamma oscillations in patients with
schizophrenia while an opposite effect was found in healthy subjects. Conclusions: rTMS
applied over the DLPFC modulates frontal gamma oscillatory in healthy subjects and in patients
with schizophrenia depending on baseline levels of activity, a finding that may ultimately
translate into a better understanding of the mechanisms leading to cognitive improvement in this
disorder.
iv
Acknowledgments
My journey to the completion of this degree involved the mentorship, collaboration, support and
love of several individuals whom I would like to acknowledge here.
It wasn‘t until I met Dr. Zafiris Jeff Daskalakis that I experienced true mentorship. Dr.
Daskalakis possesses an unwavering level of passion and dedication for research. Motivated by
his example, I attempted a similar vigorous approach to my PhD studies. I am indebted to Dr.
Daskalakis for his guidance and mentorship that resulted in a very successful and productive
PhD. More importantly, I am grateful for Dr. Daskalakis‘ positive attitude and belief in my
abilities. In addition to my insurmountable respect for Dr. Daskalakis‘ research attributes, I
respect him also for his friendships and family values. Dr. Daskalakis is a true mentor and friend
whom I will continue to consult with on both career and life endeavors.
My life changed the day I met Faranak Farzan. Faranak is probably the brightest woman I know.
She possesses a brilliant analytical mind for science that is accompanied by a nurturing heart and
soul. Faranak not only renewed my enthusiasm for research, but pushed me to analyze my data
critically. Together, commonly known as ―Meranak‖, we built a strong collaboration and an even
stronger friendship. I cherish and carry our friendship with me in my heart. Thankful I am also to
the rest of the TMS lab. I feel incredibly lucky to have been a part of such a joyful, fun and
productive lab. In particular, I would like to acknowledge Melissa Daigle and Lisa Tran for their
critical involvement in my studies. I am also grateful to Ketsa Maceus and Brenda Kirk whom
not only provided the utmost care for the patients, but also support and advise that I truly valued.
I will really miss the TMS lab. Thank you for all of the wonderful memories.
Thirty years ago, my parents Susan and David Barr, gave me the most precious and beautiful
gift. Life, opportunity, and most importantly love. Adopted from a refugee camp in Cambodia, I
was exempted from my destiny. I hope that I have made you proud. This work is dedicated to
you.
v
Table of Contents
Abstract .......................................................................................................................................... ii
Acknowledgments ........................................................................................................................ iv
Table of Contents .......................................................................................................................... v
List of Symbols and Abbreviations ............................................................................................ xi
List of Tables .............................................................................................................................. xiv
List of Figures ............................................................................................................................. xvi
List of Appendices ...................................................................................................................... xix
Chapter 1 ....................................................................................................................................... 1
1 Schizophrenia ........................................................................................................................... 1
1.1 Schizophrenia: Phenomenology ....................................................................................... 1
1.2 Schizophrenia: Epidemiology and Costs to Society ....................................................... 2
1.3 Schizophrenia: Etiology .................................................................................................... 4
1.3.1 Genetic and Environmental Risk Factors ........................................................... 4
1.4 Neurobiology of Schizophrenia ........................................................................................ 7
1.4.1 Neurochemistry ..................................................................................................... 7
1.5 Cognitive Impairment .................................................................................................... 13
1.5.1 Summary .............................................................................................................. 14
1.6 Working Memory ............................................................................................................ 15
1.6.1 Summary .............................................................................................................. 19
1.6.2 Assessment of Working Memory ....................................................................... 19
1.6.3 Working Memory: The Role of Neurotransmitter Systems ........................... 28
1.6.4 Neurophysiological Evidence for Working Memory Deficits in
Schizophrenia ...................................................................................................... 43
vi
1.6.5 Treatment for Working Memory Deficits in Schizophrenia ........................... 47
1.7 Repetitive Transcranial Magnetic Stimulation ............................................................ 49
1.7.1 Induced Cortical Changes with rTMS .............................................................. 50
1.7.2 Evidence for Working Memory Improvement in Healthy Subjects with
rTMS .................................................................................................................... 51
1.7.3 Repetitive TMS as a Treatment for Schizophrenia ......................................... 52
1.7.4 Repetitive TMS and Cortical Oscillations ........................................................ 54
1.8 Outline of Experiments ................................................................................................... 56
2 Overview of Experiments, Hypotheses and Objectives ...................................................... 57
2.1 Study 1: Evidence for Excessive Frontal Evoked Gamma Oscillatory Activity in
Schizophrenia during Working Memory ...................................................................... 57
2.1.1 Background ......................................................................................................... 57
2.1.2 Primary Objectives ............................................................................................. 58
2.1.3 Primary Hypotheses ............................................................................................ 58
2.1.4 Results .................................................................................................................. 58
2.2 Study 2: Potentiation of Gamma Oscillatory Activity through Repetitive
Transcranial Magnetic Stimulation (rTMS) of the Dorsolateral Prefrontal
Cortex ............................................................................................................................... 59
2.2.1 Background ......................................................................................................... 59
2.2.2 Primary Objective ............................................................................................... 60
2.2.3 Primary Hypothesis ............................................................................................ 60
2.2.4 Results .................................................................................................................. 60
2.3 Study 3: The Effect of Transcranial Magnetic Stimulation on Gamma
Oscillatory Activity in Schizophrenia ........................................................................... 61
2.3.1 Background ......................................................................................................... 61
2.3.2 Primary Objective ............................................................................................... 62
2.3.3 Primary Hypothesis ............................................................................................ 62
3 Study One: Evidence for Excessive Frontal Evoked Gamma Oscillatory Activity in
Schizophrenia during Working Memory ............................................................................. 63
vii
3.1 Abstract ............................................................................................................................ 63
3.2 Introduction ..................................................................................................................... 63
3.3 Materials and Methods ................................................................................................... 65
3.3.1 Subjects ................................................................................................................ 65
3.3.2 N-Back Task ........................................................................................................ 66
3.3.3 EEG Measurement of Evoked γ Oscillatory Activity ...................................... 67
3.3.4 EEG Recording ................................................................................................... 67
3.3.5 Offline EEG processing ...................................................................................... 68
3.3.6 Data Analysis ....................................................................................................... 68
3.4 Results .............................................................................................................................. 69
3.4.1 Working Memory Performance ........................................................................ 69
3.4.2 Evoked γ Power ................................................................................................... 69
3.4.3 Spectral Analysis of Other Frequency Bands .................................................. 70
3.4.4 Relationship Between Evoked Power, N-back Performance, and
Symptom Severity ............................................................................................... 70
3.4.5 Effect of Antipsychotic Medication ................................................................... 71
3.5 Discussion ......................................................................................................................... 71
4 Study 2: Potentiation of Gamma Oscillatory Activity through Repetitive
Transcranial Magnetic Stimulation (rTMS) of the Dorsolateral Prefrontal Cortex ....... 81
4.1 Abstract ............................................................................................................................ 81
4.2 Introduction ..................................................................................................................... 82
4.3 Materials and Methods ................................................................................................... 83
4.3.1 Subjects ................................................................................................................ 83
4.3.2 Procedure ............................................................................................................. 84
4.3.3 N-Back Task ........................................................................................................ 84
4.3.4 Repetitive TMS .................................................................................................... 85
4.3.5 DLPFC Site Localization .................................................................................... 86
viii
4.3.6 EEG Measurement of Evoked γ Oscillatory Activity ...................................... 86
4.3.7 EEG Recording ................................................................................................... 87
4.3.8 Offline EEG processing ...................................................................................... 87
4.3.9 Data Analysis ....................................................................................................... 87
4.4 Results .............................................................................................................................. 88
4.4.1 Evoked γ Power ................................................................................................... 88
4.4.2 EEG Spectral Analysis of Other Frequency Bands ......................................... 90
4.4.3 N-Back Working Memory Performance .......................................................... 90
4.5 Discussion ......................................................................................................................... 91
5 Study 3: The Effect of Transcranial Magnetic Stimulation on Gamma Oscillatory
Activity in Schizophrenia .................................................................................................... 104
5.1 Abstract .......................................................................................................................... 104
5.2 Introduction ................................................................................................................... 105
5.3 Materials and Methods ................................................................................................. 106
5.3.1 Subjects .............................................................................................................. 106
5.3.2 Procedure ........................................................................................................... 107
5.3.3 N-Back Task ...................................................................................................... 108
5.3.4 Repetitive TMS .................................................................................................. 108
5.3.5 DLPFC Site Localization .................................................................................. 109
5.3.6 EEG Recording ................................................................................................. 109
5.3.7 Offline EEG processing .................................................................................... 109
5.3.8 Data Analysis ..................................................................................................... 110
5.4 Results ............................................................................................................................ 111
5.4.1 N-Back Behavioural Performance ................................................................... 111
5.4.2 Evoked γ Power ................................................................................................. 111
5.4.3 EEG Spectral Analysis of Other Frequency Bands ....................................... 112
ix
5.4.4 Effect of Antipsychotic Medication ................................................................. 113
5.5 Discussion ....................................................................................................................... 113
6 Discussion .............................................................................................................................. 126
6.1 Summary of Results ...................................................................................................... 126
6.2 The Effect of rTMS on Frontal Evoked Gamma Oscillatory Activity and the
Concept of Homeostatic Plasticity ............................................................................... 128
6.3 Is there an Optimal Level of Gamma Oscillatory Activity in Working Memory? . 130
6.4 The Modulation of Gamma Activity in Working Memory Through Cortical
Inhibition ....................................................................................................................... 132
6.5 Medication Effects on Cortical Excitability and Oscillatory Activity ...................... 134
6.6 Hypo or Hyperfunctioning of the Prefrontal Cortex Contribute to Cognitive
Deficits? .......................................................................................................................... 137
6.7 Specificity of rTMS on Frontal Evoked Gamma Oscillatory Activity in the
DLPFC ........................................................................................................................... 139
6.8 Limitations and Future Directions .............................................................................. 140
6.9 Conclusion ..................................................................................................................... 143
References .................................................................................................................................. 144
7 Appendix A: Optimal TMS Coil Placement for Targeting the DLPFC Using Novel
MRI-Guided Neuronavigation ............................................................................................ 177
7.1 Abstract .......................................................................................................................... 177
7.2 Introduction ................................................................................................................... 178
7.3 Methods and Materials ................................................................................................. 180
7.3.1 Procedure ........................................................................................................... 181
7.3.2 Data Analysis ..................................................................................................... 188
7.4 RESULTS ...................................................................................................................... 189
7.4.1 Individual space ................................................................................................ 189
7.4.2 Standardized space ........................................................................................... 190
7.5 Discussion ....................................................................................................................... 197
x
Copyright Acknowledgements (if any) .................................................................................... 202
xi
List of Symbols and Abbreviations
δ Delta
θ Theta
α Alpha
γ Gamma
β Beta
5-HT 5-hydroxytryptamine
AC Anterior Commissure
ANOVA Analysis of Variance
APB Abductor Pollicis Brevis
BAs Brodmann Areas
BOLD Blood Oxygenation Level Dependent
CAMH Centre for Addiction and Mental Health
CDS Calgary Depression Scale
COMT Cathecol-O-methyltransferase
DLPFC Dorsolateral Prefrontal Cortex
DLPFCC Dorsolateral Prefrontal Cortex on the Cortex
DLPFCS Dorsolateral Prefrontal Cortex on the Scalp
DSM Diagnostic and Statistical Manual of Diseases
eDLPFCS Experimental Dorsolateral Prefrontal Cortex on the Scalp
EEG Electroencephalogram
EPSPs Excitatory Post Synaptic Potentials
ERD Event Related Desynchronization
ERS Event Related Synchronization
xii
fMRI Functional Magnetic Resonance Imaging
GABA Gamma Aminobutyric Acid
GAD 67 Glutamic Acid Decarboxylase
GAT Gamma Aminobutyric Acid Transporter
HS Healthy Subjects
IPSPs Inhibitory Post Synaptic Potentials
IQ Intelligence Quotient
MDD Major Depressive Disorder
mDLPFCS Method Dorsolateral Prefrontal Cortex on the Scalp
MEG Magnetoencephalography
MMRM Mixed Model Repeated Measures
MNI Montreal Neurological Institute
MRI Magnetic Resonance Imaging
NMDA N-Methyl-D-Asparate
NTC Non Target Correct
P Probability
PANSS Positive and Negative Symptom Scale
PAS Personality Assessment Screener
PET Positron Emission Tomography
rTMS Repetitive Transcranial Magnetic Stimulation
SANS Assessment of Negative Symptoms
SAS Statistical Analysis System
SCZ Schizophrenia
SD Standard Deviation
SE Standard Error
SICI Short Interval Cortical Inhibition
xiii
SPM Statistical Parametric Mapping
SPSS Statistical Package for the Social Sciences
TC Target Correct
tDCS Transcranial Direct Current Stimulation
TMS Transcranial Magnetic Stimulation
WM Working Memory
xiv
List of Tables
Chapter 3:
Table 1. Demographic data (±) 1 standard deviation in healthy subjects (HS) versus patients
with schizophrenia (SCZ) and the assessment of psychotic symptoms in SCZ patients………...76
Table 2. Working memory (WM) behavioural data in healthy subjects (HS) versus patients with
schizophrenia (SCZ) in the 1-, 2-, and 3-back task conditions (±) 1 standard deviation.
Behavioural data was calculated on all trials that were correct including those trials that were
rejected due to artifact…………………………………………………………………………...77
Chapter 4:
Table 1. Mean number of trials (TC + NTC) following artifact correction for each N-back
condition pre and post rTMS administration…………………………………………………….98
Table 2. Performance across N-Back task condition for target and non-target correct responses
expressed as percentage (%) before and after sham or active rTMS. Data expressed as mean (±)
standard deviation (SD)………………………………………………………………………….98
Table 3. Reaction time across N-Back task condition for target and non-target correct responses
expressed in milliseconds (msec) before and after sham or active rTMS. Data expressed as mean
(±) standard deviation (SD)………………………………………………………………………99
Chapter 5:
Table 1. Demographic Data for Healthy Subjects (HS) and patients with schizophrenia (SCZ)
and the assessment of psychotic symptoms in patients with SCZ rTMS (±) 1 standard
deviation………………………………………………………………………………………...119
Table 2. Total number (TC+NTC) of trials analyzed for healthy subjects (HS) and patients with
schizophrenia (SCZ) in the 1-, 2-, and 3-back task conditions pre- post-rTMS (±) 1 standard
deviation………………………………………………………………………………………...120
Table 3. Working memory (WM) behavioural performance (%) and reaction time (RT; msec)
during the N-back in healthy subjects (HS) versus patients with schizophrenia (SCZ) pre-post
either active or sham rTMS (±) 1 standard deviation pre-post rTMS…………………………..121
Appendix A:
Table 1: Actual distance (cm) from the experimental location of left DLPFC (eDLPFCS) to EEG
electrodes, APB, and the ‗5-cm method‘ on the scalp in all 15 healthy subjects in individual
space…………………………………………………………………………………………….193
xv
Table 2: A comparison showing the voxel coordinates predicted by our method (mDLPFCS)
with the voxel coordinates obtained during the experiment with the co-registration probe
(eDLPFCS) in individual space…………………………………………………………..……..193
Table 3: Positions in standard MNI space (mm) for each subject predicted by the method
(mDLPFCS) compared to positions measured during the experiment (eDLPFCS)……………..194
xvi
List of Figures
Chapter 3:
Figure 1. A) A representation of the 1-, 2- and 3-back conditions that were completed in a
randomized order by patients with schizophrenia (SCZ) and healthy subjects. Subjects were
required to push one button (target) if the current letter was identical to the letter presented ―N‖
trials back; otherwise the participants pushed a different button (non-target). Correct responses
for target (TC) and non-target (NTC) were included in the data analysis. B) The timing of one
trial from the presentation of a one letter separated by a (+) sign followed by a subsequent letter
for a total time of 3000 msec…………………………………………………………………….78
Figure 2. A) Mean log transformed γ (30-50 Hz) oscillatory power (TC+NTC) elicited across N-
back task conditions in patients with schizophrenia (SCZ) compared to healthy subjects (HS).
Error bars represent (±) 1 standard error (SE). B) Topographical illustration of mean absolute γ
power elicited during 3-back in patients with SCZ compared to HSs. Greater γ power is
represented by hot colours………………………………………………….……………………79
Figure 3. Mean log transformed frequency power in the δ (1-3.5 Hz), (4-7 Hz), (8-12 Hz),
and (12.5-28 Hz) ranges (TC+NTC) elicited across N-back task conditions in patients with
schizophrenia (SCZ) compared to healthy subjects (HS). Error bars represent (±) 1 standard error
(SE)………………………………………………………………………….…………………...80
Chapter 4:
Figure 1. N-Back WM Task. A) A representation of the four different WM Conditions (0, 1, 2,
and 3-back) that were administered in a randomized order to participants before and after rTMS.
B) The timing of one trial from the presentation of a one letter separated by a (+) sign followed
by a subsequent letter for a total time of 3000 msec. Participants were required to push one
button (target) if the present letter was identical to the letter presented ―N‖ trials back; otherwise
the participants pushed a different button (non-target)…………………………………………100
Figure 2. Targeting the Dorsolateral Prefrontal Cortex for rTMS Administration. Transverse
view from a single subject with exposed cortex and overlap of Brodmann areas 9 & 46,
highlighted (white) on a T1-weighted 3D MRI. Using MRI-to-MiniBird co-registration, the
centre of the TMS coil was held over this region………………………………………………101
Figure 3. Mean log transformed change in oscillatory power (post rTMS frequency power-pre
rTMS frequency power) following rTMS elicited during the N-back task for A) γ power
compared to δ, , , and activities following B) sham, and C) active stimulation. Error bars
represent (+) standard deviation (SD)…………………………………………………………..102
Figure 4. A) Topographical illustration of mean absolute change (post rTMS γ power-pre rTMS
γ power) in γ power elicited during 3-back condition following sham and active rTMS. Maximal
change in γ power (represented by hot colours) was found following active stimulation in the
frontal brain regions. B) Change in the mean sum of γ power in the frontal
(FPZ+FP1+FP2+AF3+AF4) versus posterior (OZ+O1+O2+PO3+PO4) electrodes in the 3-back
xvii
condition following sham and active stimulation. Error bars represent (+) standard deviation
(SD)………………………………………………………………………………….………….103
Chapter 5:
Figure 1. A) A representation of the 1-, 2- and 3-back conditions that were completed in a
randomized order by patients with schizophrenia (SCZ) and healthy subjects (HS) pre-post
rTMS. Subjects were required to push one button (target) if the current letter was identical to the
letter presented ―N‖ trials back; otherwise the participants pushed a different button (non-target).
Correct responses for target (TC) and non-target (NTC) were included in the data analysis. B)
The timing of one trial from the presentation of a one letter separated by a (+) sign followed by a
subsequent letter for a total time of 3000 msec………………………………………………...122
Figure 2. Targeting the Dorsolateral Prefrontal Cortex (DLPFC) for rTMS stimulation.
Transverse view from a single subject with exposed cortex and overlap of Brodmann areas 9 &
46, highlighted (white) on a T1-weighted 3D MRI. Using MRI-to-MiniBird co-registration, the
centre of the TMS coil was held over this region………………………………………………123
Figure 3. A) Mean log transformed gamma oscillatory power (γ; 30-50 Hz; uV2) for target
correct (TC) and non-target correct (NTC) responses during the N-back task pre-post rTMS in
healthy subjects (HS; N=22) versus patients with schizophrenia (SCZ; N=24). B) Mean log
transformed γ oscillatory power (uV2) for target correct (TC) and non-target correct (NTC)
responses during the 3-back condition measured from the frontal and posterior brain regions pre-
post rTMS in patients with schizophrenia (SCZ; N=24) and healthy subjects (HS; N=22). Bars
represent (±) 1 standard deviation………………………………………………………………124
Figure 4. Mean log transformed oscillatory power (uV2) for target correct (TC) and non-target
correct (NTC) responses during the N-back task pre-post rTMS in healthy subjects (HS; N=22)
versus patients with schizophrenia (SCZ; N=24) across delta (δ; 1-3.5 Hz), theta (; 4-7 Hz),
alpha (; 8-12 Hz), and beta (; 12.5-28 Hz) frequency ranges. Bars represent (±) 1 standard
deviation………………………………………………………………………………………...125
Appendix A:
Figure 1. A non-linear transformation to convert the coordinates from a standard space to an
individual space was first estimated (Step 1A). This estimated transform was then applied to the
coordinates for the DLPFC in the standard brain (or template) to determine the position of the
DLPFC in the individual brain (Step 1B). Next, a triangular mesh wrapping the iso-surface
representing the scalp was created with the marching cube algorithm (see Methods). The optimal
position for the coil (mDLPFCS) on the scalp was then defined as the vertex of the triangular
mesh with a normal that passed through the DLPFC on the cortex (DLPFCC; Step 2)………...183
Figure 2. A) In this coronal slice of a MRI image, the orange spot represents the DLPFC on the
cortex (DLPFCC) and the green squares are voxels on the iso-surface with a normal that passes
through the DLPFCC in a distance less than 3 cm. Voxels (green squares) on surfaces other than
the scalp (i.e., the internal side of the skull, and the interface gray matter-CSF) were filtered out
if they had a normal greater than 25° from the average normal. The mDLPFCS was defined as the
intersection of the scalp with a line with the orientation of the average normal and passes through
their average position. B) The position of mDLPFCS was superimposed on each individual's MRI
xviii
for later co-registration. Once mDLPFCS is co-registered, this position is referred to as the
eDLPFCS………………………………………………………………………………………..186
Figure 3. Although the position predicted by our method (mDLPFCS; green) is placed on the
subjects‘ scalp, the actual position measured with the probe when co-registered (eDLPFCS;
orange) is biased in the radial direction. These positions are shown for all subjects in MNI
space…………………………………………………………………………………………….192
Figure 4. The dispersion of APB, ‗5-cm method‘, eDLPFCS and electrode positions are
represented by spheroids in standardized space. Spheroids are centered at the mean position for
each measurement with a radius in each direction (i.e. X, Y, and Z) equal to 1 standard deviation.
Please note that these positions were measured on the scalp so they are ―flying‖ over the gray
matter…………………………………………………………………………………………...196
xix
List of Appendices
A: Optimal Identification of the DLPFC using MRI-Guided Neuronavigation………………177
1
Chapter 1
1 Schizophrenia
1.1 Schizophrenia: Phenomenology
Schizophrenia is still one of the most serious and debilitating mental disorders of modern
medicine. The classification of schizophrenia has a long history that ranges from narrow to broad
definitions based on both abstract theories and empirical data. Schizophrenia was first classified
by Bendict Morel in 1853. Morel described schizophrenia as a syndrome which affects young
adults which he referred to as démence precoce. The classification of psychoses was later
significantly advanced by Emil Kraepelin (1856-1926) and is still influential today. In 1898, he
proposed the term dementia praecox (premature dementia) for schizophrenia based on
longitudinal studies of the patients‘ conditions over time. In Kraepelin‘s description, dementia
praecox has an early onset that progressively deteriorates to a common end stage. Despite the
dismal outcome described by Kraepelin, he acknowledged the need for cross-sectional diagnostic
guidelines including bizarre thought disturbances, delusions, auditory or other hallucinations,
catatonia, abnormalities in volition and ―a pervasive reduction in cognitive and affective
capacity‖, in addition to, ―peculiar states of mental weakness‖. Importantly, Kraepelin believed
that patients with dementia praecox do not lose metal abilities but rather lose the ability to use
them appropriately. Kraepelin‘s definition of schizophrenia was later challenged by Eugen
Bleuler (1857-1939) for being too narrow and described only the severe cases. Instead Bleuler‘s
definition included a more heterogeneous grouping that emphasized the loss of mental
connectedness in which then he came to name this disease as schizophrenia. According to
2
Bleuler, associative splitting is present in the fundamental symptoms of schizophrenia that
included: (1) association defects; (2) autism; (3) ambivalence; and (4) disturbance of affect.
Bleuler‘s categorization of schizophrenia was widely accepted and used by clinicians until the
publication of the Diagnostic and Statistical Manual of Diseases (DSM-II) in 1980. According to
the fourth edition of the DSM; there are five subtypes of schizophrenia: 1) Paranoid; 2)
Disorganized; 3) Catatonic; 4) Undifferentiated; and 5) Residual. Despite the widespread use of
the DSM-IV as a diagnostic tool, scientists and clinicians remain skeptical of disease categories
and are in favour of a more mechanistic basis of brain diseases.
Schizophrenia is characterized by three main categories of symptoms: positive and negative
symptoms and cognitive deficits. Positive symptoms include auditory hallucinations, bizarre
delusions and thought disorder, while negative symptoms include blunted affect, alogia, and
anhedonia. Cognitive deficits in patients with schizophrenia have been shown across a wide
variety of cognitive tasks including executive functioning, attention, memory, working memory
and general intellectual functioning (Weickert, Goldberg et al. 2000). Antipsychotic medications
are effective in managing positive symptoms however little improvement has been shown on
negative symptoms and cognitive deficits.
1.2 Schizophrenia: Epidemiology and Costs to Society
A recent review of epidemiological data suggests that the lifelong prevalence and incidence of
schizophrenia is between 0.30-0.66% and 10.2-22.0 per 100 000 persons per year (McGrath,
Saha et al. 2008). In Canada, the estimated number of persons with schizophrenia was 234 305 in
2004 (Goeree, Farahati et al. 2005). However these numbers depend on the criteria used to define
schizophrenia (Castle, Wessely et al. 1993). Symptoms typically begin during late adolescence
and may differ between genders. That is, there are reports of males developing symptoms earlier
3
than females although this has not been consistently shown across countries and at all ages of
onset (Jablensky 2000). The outcome of schizophrenia has been shown to improve with follow
up care by approximately 60% (Harding, Brooks et al. 1987) and reduce relapses through
combined antipsychotic medication and psychosocial rehabilitation (Hogarty 1993).
Furthermore, studies have shown that outcome in industrialized countries is typically worse as
compared to third world countries (Kulhara and Chandiramani 1988; Ohaeri 1993). For example,
according to a World Health Organization study of 10 countries, significantly better outcomes
were found in Nigeria, India and Colombia compared to the United States, Japan and England
(Jablensky 2000).
Financial costs to society are substantial due to the chronic nature of schizophrenia and its early
onset. Globally, the direct costs of schizophrenia on healthcare budgets for the costs of
prescription medications, hospitalizations, diagnoses, procedures and long-term care services are
approximately 1.5-3% of the total national healthcare expenditure (Knapp 1997; Wu, Birnbaum
et al. 2002; Andlin-Sobocki and Rossler 2005; Chisholm, Gureje et al. 2008). The indirect costs
of schizophrenia are even higher than the direct costs including increased unemployment,
reduced workplace productivity, caregiver burden, and premature mortality from suicide (Knapp
1997; Goeree, Farahati et al. 2005; Chang, Cho et al. 2008). Patients with schizophrenia also
have high rates of physical comorbidity including diabetes, hyperlipidaemia, cardiovascular
disease and obesity further substantiating these costs (Lambert, Velakoulis et al. 2003). In the
United States, the overall cost of schizophrenia was estimated to be $62.7 billion in 2002 of
which $22.7 billion comprised of direct cost in addition to $7.6 billion of indirect cost (Wu,
Birnbaum et al. 2002). In the United Kingdom, £810 million was incurred by persons with
schizophrenia equaling 2.76% of total National Health Service expenditure in 1992-1993 (Knapp
1997). In Canada, the burden of schizophrenia on health care was estimated to be $2.02 billion in
4
2004. Additionally costs of employment rate, productivity morbidity, and mortality loss was
$4.83 billion for the total cost of schizophrenia of $6.85 billion in 2004 (Goeree, Farahati et al.
2005). Finally, the landmark World Bank and Wold Health Organization study The Global
Burden of Disease, ranked schizophrenia 5th
among the leading causes of disability measured by
Disability-Adjusted Life Years in industrialized countries and 9th
worldwide (Murray and Lopez
1996).
1.3 Schizophrenia: Etiology
The pathophysiology underlying schizophrenia is still not completely understood. An interaction
between genetic predisposition and several environmental risk factors however has been
suggested to contribute to the development of this disease.
1.3.1 Genetic and Environmental Risk Factors
Schizophrenia is a familial disease despite the fact that more than eighty percent of persons with
schizophrenia have no affected relatives within the first degree of kinship and more than sixty
percent have families without a history of this disorder (Nasrallah and Smeltzer 2002). Studies in
monozygotic twins indicate that the concordance rate is approximately fifty percent (Kaplan,
Sadock et al. 1994). In addition, family and twin studies suggest that in first-degree relatives and
dizygotic twins the risk for developing schizophrenia is estimated to be ten-fifteen percent
(Kendler and Robinette 1983; McGuffin, Farmer et al. 1984; Kaplan, Sadock et al. 1994).
Adoption studies have also yielded interesting results showing an increased risk of developing
schizophrenia in children of schizophrenic birth parents adopted into non-schizophrenic homes
(Heston 1966); while children with non-schizophrenic parents adopted into schizophrenic
families do not (Wender, Rosenthal et al. 1977).
5
Over the past twenty years, identification of genetic linkage using DNA polymorphism
technology (Botstein, White et al. 1980) has now been applied to schizophrenia. In this regard,
linkage studies have identified loci on chromosomes 1, 2, 4, 5, 6, 7, 8, 9, 10, 13, 15, 18, 22, and
X implicated at the Sixth World Congress on Psychiatric Genetics (Brzustowicz, Hodgkinson et
al. 2000) (Barden and Morissette 1999; Craddock and Lendon 1999; Crowe and Vieland 1999;
Curtis 1999; Detera-Wadleigh 1999; Gejman 1999; Hallmayer 1999; Kennedy, Basile et al.
1999; Nurnberger and Foroud 1999; Paterson 1999; Schwab and Wildenauer 1999; Van
Broeckhoven and Verheyen 1999; Wildenauer and Schwab 1999; Riley and McGuffin 2000).
However, such findings should be replicated as these studies have produced conflicted reports or
did not satisfy statistical criteria for significant linkage (Lander and Kruglyak 1995). These
studies though identified such chromosome loci by genome scanning methods that were not
hypothesis driven. Although there is a clear association between altered dopamine activity in
schizophrenia, studies have failed to establish a link between dopamine receptors or transporter
genes with this disorder (Coon, Jensen et al. 1994; Hallmayer, Maier et al. 1994; Maier, Schwab
et al. 1994; Nanko, Fukuda et al. 1994; Ravindranathan, Coon et al. 1994). There has been
however reports of variations in the cathecol-O-methyltransferase (COMT) gene (Egan,
Goldberg et al. 2001) and varied reports with 5-hydroxytryptamine (serotonin; 5-HT) receptor
genes (Williams, Spurlock et al. 1996; Owen 2000) in the frontal cortex associated with
increased risk for the development of schizophrenia.
An area on chromosome 22q11-13 has been linked to psychosis and schizophrenia. More
specifically, abnormalities in chromosomes 5q, 11q, 18q, 19q, and 22q have been associated with
both psychosis and schizophrenia (Bassett 1992). The 22q11.2 area is subject to deletions that
have been associated with the DiGeorge and velocardiofacial syndromes (Amati, Conti et al.
1999) and increased risk of schizophrenia (Bassett, Chow et al. 2000). DISC-1 is another gene
6
that has been shown to be disrupted in patients with schizophrenia in Scottish (Millar, Wilson-
Annan et al. 2000; Millar, Christie et al. 2001) and Finnish (Ekelund, Hovatta et al. 2001)
families although the function of this gene has not been well characterized. To date,
chromosomal and gene abnormalities associated with schizophrenia have yielded some
interesting results, however, these findings are not selectively linked to schizophrenia as they
overlap with other psychiatric disorders. Moreover, current diagnoses have yet to establish
biological validity or validation and the identification of phenotypes have not been simplified by
genetic association (Bertolino and Blasi 2009).
In addition to the genetic predisposition of schizophrenia, several environment risk factors have
been identified that contribute to the development of this disorder. Moreover, the time in which
individuals are exposed to such environmental risk factors is important. For example,
complications during pregnancy, reduced vitamin D levels, increased stress to the mother,
diabetic mother, and those mothers older than 35 years of age during the prenatal stage have
been shown to contribute to the development of schizophrenia (McGrath 1999; Cannon, Jones et
al. 2002; Wohl and Gorwood 2007; Khashan, Abel et al. 2008). Post-natal environmental risk
factors such as fetal hypoxia and babies born at higher altitudes have been reported to increase
the probability of developing schizophrenia (Krabbendam and van Os 2005; McGrath, Saha et al.
2008). Exposure to other environmental risk factors such as ethnicity, urbanization, and cannabis
use during childhood to young adulthood has also been identified in the development of
schizophrenia (Krabbendam and van Os 2005; McGrath, Saha et al. 2008; Veling, Mackenbach
et al. 2008).
The search for endophenotypes of interest for specific genetic factors in schizophrenia has been
considered. Notably, cognitive deficits may represent endophenotypes which are intermediate
7
phenotypes that have been suggested to provide a more reliable index of liability than the illness
itself (Gottesman and Shields 1972). In this regard, a meta-analysis performed on unaffected
first-degree relatives of schizophrenia patients compared to healthy subjects demonstrated
significant group differences in performance across attention/working memory, verbal memory,
visual memory, executive function, spatial ability, motor function, language function, and
general intelligence cognitive domains (Snitz, Macdonald et al. 2006). This study in addition to
previous meta-analyses (Grove, Lebow et al. 1991; Cannon, Zorrilla et al. 1994; Park, Holzman
et al. 1995; Faraone, Seidman et al. 1996; Cosway, Byrne et al. 2000; Curtis, Calkins et al. 2001;
Appels, Sitskoorn et al. 2003) demonstrate that cognitive deficits, notably in executive control
functions, may serve as valuable endophenotypes of interest in search for specific genetic factors
related to schizophrenia.
1.4 Neurobiology of Schizophrenia
1.4.1 Neurochemistry
1.4.1.1 Dopamine
Over the last 5 decades, the involvement of dopamine in the pathophysiology and treatment of
schizophrenia has been an area of immense research. Dopamine activates 5 types of
metabotropic G protein-coupled receptors in the brain: D1, D2, D3, D4 and D5 (Neve, Seamans
et al. 2004). Dopamine is present in the ventral tegmental area of the midbrain, substantia nigra,
and hypothalamus. Moreover, dopamine is involved in the control of information flow in the
frontal lobe to other areas of the brain. Dopamine serves several important functions in the brain
including motivation, sleep, mood, voluntary movement and working memory (Jones, Kilpatrick
et al. 1986). The dopamine hypothesis first proposed that hyperactive dopamine
neurotransmission underlies the presentation of positive symptoms of schizophrenia (Carlsson
8
and Lindqvist 1963). This hypothesis was based on the relationship between the dose of
antipsychotic medication and their potency to block dopamine D2 receptors (Seeman and Lee
1975; Creese, Burt et al. 1976) and also by the psychotogenic effects of dopamine-enhancing
drugs (Angrist and van Kammen 1984; Lieberman, Kane et al. 1987). This classical formulation
of the dopamine hypothesis mostly emphasized subcortical regions including the striatum and the
nucleus accumbens based on the prevalence of dopamine terminals and D2 receptors.
Although the classical dopamine hypothesis satisfied the relationship between dopamine and
positive symptoms, it did not account for persistent negative and cognitive symptoms with
antipsychotic treatment. Moreover, the classical dopamine hypothesis did not explain the
relationship between genetics, neurodevelopment deficits or other known risk factors for
schizophrenia (Howes and Kapur 2009). In this regard Davis et al (1991) published a seminal
paper on the reformulation of the classical dopamine hypothesis that was based on more recent
evidence from neuroimaging, post-mortem, and animal studies, which placed the emphasis on
the interaction between hypodopaminergic activity in the prefrontal cortex and
hyperdopaminergic activity in more subcortical regions (Davis, Kahn et al. 1991). For example,
neuroimaging studies have suggested that an alteration in prefrontal cortical functioning may
underlie negative and cognitive symptoms in this disorder (Knable and Weinberger 1997).
Evidence from preclinical studies has demonstrated the importance of dopamine
neurotransmission at D1 receptors, which is the predominant dopamine receptor in the neocortex,
in the proper functioning of the prefrontal cortex (Goldman-Rakic, Muly et al. 2000). In humans
it has been shown that carriers of high-activity allele of COMT, an enzyme involved in the
metabolism of dopamine, perform worse on cognitive tasks compared to those carriers of the
allele that induces lower concentration of dopamine in the prefrontal cortex (Goldberg and
Weinberger 2004). Finally, clinical studies also support this association revealing a relationship
9
between reduced cerebrospinal fluid homovanillic acid with indexes low dopamine activity in the
prefrontal cortex and poor performance on working memory tasks (Weinberger, Berman et al.
1988; Kahn, Harvey et al. 1994). This modified dopamine hypothesis therefore proposed that
positive symptoms arise from striatal hyperdopaminergia while prefrontal hypodopaminergia
may underlie negative and cognitive symptoms in schizophrenia.
The second reconceptualization of the dopamine hypothesis was criticized for its lack of direct
evidence for reduced dopamine in the frontal cortex and limited direct support for hyperactive
dopamine in the striatum (Howes and Kapur 2009). In this regard, Howes and Kapur recently
published a third version of the dopamine hypothesis in 2009 that accounts for recent evidence of
known risk factors in addition to sociocultural factors that together contribute to schizophrenia
(Howes and Kapur 2009). This hypothesis proposed 4 distinctive components. The first
component proposes that several ―hits‖ including fronto-temporal dysfunction, genes, stress, and
drugs contribute to dopamine dysregulation referred to as the final pathway to psychosis in this
disorder. The second component places less emphasis on the D2 receptor level to the level of
presynaptic dopaminergic control level contributing to the dopamine dysregulation. The third
component relates dopamine dsyregulation to psychosis rather than leading to schizophrenia
itself. That is, the specific diagnosis is a result of the interaction of hits that interact with
sociocultural contributions. The final fourth component hypothesizes that dopamine
dsyregulation alters the evaluation of stimuli possibly through aberrant salience (Howes and
Kapur 2009).
1.4.1.2 Serotonin
Serotonin has also been examined in schizophrenia and is targeted in newer antipsychotic
medication. Serotonin neurotransmission has been implicated in the regulation of mood and
10
suicidal tendencies across neuropsychiatric disorders (Bellivier, Szoke et al. 2000). Serotonin is
released predominantly from the raphe nuclei in the brain stem and is produced from the amino
acid L-tryptophan by tryptophan hydroxylases2 (Grohmann and Trendelenburg 1988). In the
lower raphe nuclei, axons innervate the cerebellum and the spinal cord while the axons from the
upper raphe nuclei innervate the entire brain (Grohmann and Trendelenburg 1988). There are at
least 14 different subtypes of serotonin receptors that have been further classified into 7
subfamilies of which are mostly part of the G-protein linked receptors that exert both excitatory
and inhibitory cell responses (Wong and Van Tol 2003). Serotonin receptor activity involves the
signaling through several different mechanisms including Gi/o which inhibit adenylyl cyclase,
PCL-, and adenylyl cyclase and is expressed in many different peripheral tissues as well as the
central nervous system (Raymond, Mukhin et al. 2001). An alteration in serotonin activity in
schizophrenia was first proposed owing to the evidence of its similar chemical structure with the
hallucinogenic drug lysergic acid diethylamide (LSD; (Gaddum 1954); (Wooley and Shaw
1954)). This has since been supported by post-mortem and neuroimaging studies which have
reported an alteration in serotonin activity in patients with schizophrenia. For example, post-
mortem studies consistently demonstrate increased levels of 5-HT1A receptor subtypes examined
through 5-HT1A receptor agonist [3H]8-hydroxy-2-[di-n-propyl-amino]tetralin([
3H]8-OH-DPAT)
in the prefrontal cortices of patients with schizophrenia (Hashimoto, Nishino et al. 1991; Joyce,
Shane et al. 1993; Simpson, Lubman et al. 1996; Sumiyoshi, Stockmeier et al. 1996; Gurevich
and Joyce 1997). Reductions in the density of 5-HT transporters labeled with [3H]paroxetine
(Laruelle, Baldwin et al. 1993), [3H]cyanoimipramine (Joyce, Shane et al. 1993), and with
[3H]citalopram (Ohuoha, Hyde et al. 1993) have also been shown through
magnetoencephalography (MEG) in the frontal cortices of patients with schizophrenia compared
to healthy subjects and other brain regions. Decreases in 5-HT2A density have also been reported
11
although these findings may be confounded by medication effects. Together these studies
suggest an alteration in serotonin activity in the prefrontal cortex of patients with schizophrenia
however such abnormalities are not as influential as the evidence for an impaired dopamine
system in this disorder.
1.4.1.3 Glutamate
Excitatory neurotransmission is predominantly mediated by the neurotransmitter glutamate.
Glutamate is involved in several cognitive functions such as learning, memory, and synaptic
plasticity (Johnson 1972). Glutamate is synthesized locally or as a product of the Krebs cycle and
binds to both ionotropic and metabotropic receptors. For example, glutaminergic excitatory
neurotransmission is mediated by three main types of ionotropic receptors: N-methyl-D-asparate
(NMDA), amino-hydroxy-5-methyl-4-isoxazole propionic acid and kainic acid (Newcomer and
Krystal 2001). The glutamate hypothesis of schizophrenia was based on the finding of increased
psychotic symptoms in healthy subjects following the channel blocking of NMDA receptors and
the exacerbation of positive and cognitive symptoms in patients with schizophrenia (Luby,
Cohen et al. 1959; Tamminga 1998; Krystal, D'Souza et al. 1999). Animal studies have
demonstrated that reduced levels of NMDA receptor subunit results in behavioural changes
similar to symptoms of schizophrenia including hyper locomotion and social withdrawal in mice.
Moreover, such induced behavioural changes were ameliorated by antipsychotics (Mohn,
Gainetdinov et al. 1999). In addition, evidence of glutamate hypofunctioning has also been
revealed through neuroimaging studies in patients with schizophrenia during rest or while
performing cognitive tasks. More recent studies then suggested that the noncompetitive NMDA
receptor antagonists were found to preferentially increase metabolism and extracellular
glutamate levels localized in the limbic circuits rather than throughout the entire cortex (Lahti,
Koffel et al. 1995; Moghaddam, Adams et al. 1997; Duncan, Miyamoto et al. 1999; Holcomb,
12
Lahti et al. 2005). Although these studies do not suggest a generalized reduction in NMDA
receptor activity throughout the entire brain, it is clear that impaired NMDA function that is
critical to limbic forebrain functioning is involved in the pathophysiology of schizophrenia
(Marek, Behl et al.). Furthermore, genetic studies have failed to find a link between
schizophrenia and NMDA genes. Finally, glutamate is also involved in inhibitory
neurotransmission serving as a precursor in the synthesis of gamma-aminobutyric acid (GABA),
a mechanism that has also been shown to be impaired in patients with schizophrenia and will be
reviewed in the following section.
1.4.1.4 GABA
Altered inhibitory neurotransmission has also been implicated in the pathophysiology of
schizophrenia (Roberts and Frankel 1950). Inhibitory neurotransmission is a process that
involves the suppression of cortical activity and is predominantly mediated by GABA that is
present in all cortical layers of the brain (Jones 1993). GABA is almost exclusively synthesized
by the excitatory neurotransmitter glutamate and its effects are mediated by ionotropic GABAA
and metabotropic GABAB receptors. GABAA receptor activity is reliant on the concentration of
cation-chloride while activity of GABAB receptors depends on the presence of calcium and
magnesium. GABAB receptors are located mainly in the presynaptic membrane of aminergic
synapses and when GABA or GABA agonists bind to these receptors, the release of dopamine,
noradrenergic, and serotonin is inhibited. The GABAergic hypothesis of schizophrenia was
introduced shortly after discovery of GABA and its inhibitory role in the central nervous system
(Roberts and Frankel 1950; Roberts 1972).
Impaired GABAergic inhibitory neurotransmission in schizophrenia has been specifically shown
in the dorsolateral prefrontal cortex (DLPFC). For example, post-mortem studies report
13
reductions of the messenger RNA for 67 kiloDalton isoform of glutamic acid decarboxylase
(GAD 67), an enzyme that synthesizes GABA, in 25-30% of GABAergic interneurons in the
DLPFC (Akbarian, Kim et al. 1995; Volk, Pierri et al. 2002; Hashimoto, Volk et al. 2003).
Furthermore, decreased densities of interneurons in layer II of the prefrontal cortex of
schizophrenia patients has also been reported (Benes, McSparren et al. 1991). Reductions in the
synthesis of GABA have also been demonstrated in GABAergic neurons that express calcium
binding protein parvalbumin in the DLPFC of patients with schizophrenia (Lewis, Hashimoto et
al. 2005). Such deficits in GABAergic activity in patients with schizophrenia has been suggested
to result from excessive activation of dopamine D2 receptors resulting in heightened excitability
in the cortex possibly leading to psychotic symptoms (Benes 1997).
1.5 Cognitive Impairment
Cognitive deficits are now considered a core component in schizophrenia (Elvevag and Goldberg
2000). Deficits in cognition are estimated to occur in 75-85% of patients with schizophrenia and
often are presented prior to the onset of other symptoms (Reichenberg, Weiser et al. 2006).
Despite the effectiveness of antipsychotic medication for the management of positive symptoms,
cognitive impairment persist (Heinrichs 2005). Cognitive impairment has been shown as the best
predictor of long term functional outcome compared to indices of the positive or negative
symptom domain (Green 1996; Gold 2004). Relationships have been demonstrated, though,
between cognitive performance and negative symptoms (Harvey, Howanitz et al. 1998; Park,
Puschel et al. 1999). Moreover, it has been argued that schizophrenia patients with the greatest
cognitive impairment present the most prominent negative symptoms (Basso, Nasrallah et al.
1998; Villalta-Gil, Vilaplana et al. 2006). It has further been suggested that negative symptoms
14
mediate the relationship between cognition and functional outcome (Ventura, Hellemann et al.
2009).
Patients with schizophrenia generally perform one standard deviation below the mean on tests of
cognition (Weickert, Goldberg et al. 2000; Fioravanti, Carlone et al. 2005). A meta-analysis
conducted by Fioravanti et al 2005 analyzed studies that examined cognitive functioning across
the traditional domains of cognition that includes the assessment of: 1) global cognitive
functioning or IQ; 2) memory functioning; 3) language; 4) executive functioning; and 5)
attention in patients with schizophrenia compared to healthy subjects. The results of the meta-
analysis revealed that patients with schizophrenia perform worse than healthy subjects across all
cognitive domains (Fioravanti, Carlone et al. 2005). Moreover, the greatest significant
differences between the two groups were found in memory, language and IQ. Fioravanti et al
(2005) therefore confirming previous studies (i.e., (Heinrichs and Zakzanis 1998) demonstrating
a generalized cognitive deficit in patients with schizophrenia compared to healthy subjects.
Importantly this study acknowledges sources of heterogeneity such as patient characteristics (i.e.,
duration of illness, medication) and also generic differences such as the lack of motivation.
Despite these factors that could have contributed to within patient group differences,
nevertheless, it is evident from studies conducted from 1990-2003 that patients with
schizophrenia are impaired across all cognitive domains (Fioravanti, Carlone et al. 2005).
1.5.1 Summary
Schizophrenia is a complex neuropsychiatric disorder that presents positive, negative, and
cognitive deficits as part of its symptom profile. These symptoms have been linked to
abnormalities in dopamine, GABA, and other neurotransmitter systems. Furthermore, there is
considerable evidence for a genetic component in schizophrenia however interactions with
15
external environmental risk factors have also been identified. Cognitive impairment in particular
has been demonstrated as the best predictor of long term functional outcome and therefore such
deficits are well studied in this disorder. Working memory is a higher-order cognitive process
that involves the maintenance and manipulation of information on line (Baddeley 1986). This
process is involved in many cognitive tasks and is important in everyday functioning. There is
considerable evidence for working memory impairment in patients with schizophrenia therefore
underscoring the need for a better understanding of the mechanisms underlying this cognitive
process. The proceeding sections will review this literature on the underlying mechanisms of
working memory and evidence for impaired working memory function in patients with
schizophrenia.
1.6 Working Memory
Working memory is commonly defined as the ability to temporary retain information that no
longer exists in the external environment or is retrieved from long-term memory (D'Esposito
2007). Such internal representations are short-lived, but can be stored for longer periods of time
through the active maintenance or rehearsal strategies, and can be also subjected to various
operations that manipulate the information for goal-directed behaviour (D'Esposito 2007).
Working memory is involved in everyday functioning such as calculating the tip amount at a
restaurant and is critical to other cognitive abilities such as reasoning, language comprehension,
planning and spatial processing (D'Esposito 2007). To this end, working memory has been
proclaimed as ―perhaps the most significant achievement of human mental evolution‖ (Goldman-
Rakic 1992).
Working memory is a theoretical construct that is used in the fields of cognitive psychology and
cognitive neuroscience and each discipline has defined working memory through several
16
different models (for review, (Miyake and Shah 1999)). The first and most influential proposal of
working memory was put forth by Baddeley and Hitch in 1974 that involved multiple
components to mediate executive control and the active maintenance of temporarily stored
information. In Baddeley‘s working memory model, four components were proposed in which
two of these components serve as storage buffers for the maintenance of visual and verbal
information referred to as the visuo-spatial sketchpad and the phonological loop (Baddeley and
Hitch 1974). A third episodic buffer was then included in this model for the storage of
information in a multi-dimensional code and serves as an interface between the visual and verbal
buffers and the fourth component referred to the central executive component (Baddeley 2000).
The central executive component is then hypothesized to guide the manipulation and
transformation of information held within the storage buffers (Baddeley and Hitch 1974).
A critical component of Baddeley‘s working memory model is the existence of verbal and spatial
storage buffers. Moreover, the cognitive concept of a buffer indicates that the temporary storage
of verbal and spatial information requires the transfer of information from one brain region to
another (D'Esposito 2007). In this regard, a wealth of studies investigating the neural correlates
of working memory consistently report the involvement of a number of cortical regions,
particularly regions within the prefrontal cortex, anterior cingulate area, and posterior parietal
cortex (Jonides, Schumacher et al. 1998; Fletcher and Henson 2001; Leung, Gore et al. 2002;
Sakai, Rowe et al. 2002). It has been shown that different brain regions are activated while
performing working memory tasks that are dependent on the task and therefore may be related to
task-specific processing or the modalities in which the task employs. However, among these
brain regions the DLPFC (Brodmann areas (BAs) 9, 46, and 9/46) is consistently shown to be
activated across working memory tasks and has been considered for the role of the central
executive. Moreover, the central executive is believed to be involved in the manipulation of
17
information being stored in the domain-specific buffers, sustained attention during competing
information, temporal coding or sequencing, updating the contents of working memory, and the
maintenance of goal representations in working memory (Barch and Smith 2008). Other brain
regions however have also been implicated to play the role of the central executive in the
development of more recent working memory theories.
There are several competing working memory theories that have been proposed since the time of
Baddeley with the advancement in neuroimaging. For example, Wager and Smith (2003)
conducted a meta-analysis on neuroimaging working memory studies that recruited a central
executive showed activation in the posterior parietal cortex (Wager and Smith 2003). By
contrast, Jonides et al (1998) demonstrated increased processing in the left prefrontal cortex
while performing a verbal working memory task measured with positron emission tomography
(PET) in healthy subjects (Jonides, Schumacher et al. 1998). Such inconsistent results may be
attributed to the type of working memory or the level of task difficulty tested. In this regard, the
involvement of the DLPFC and ventrolateral prefrontal cortex in the mediation of working
memory across different types of information being processed (i.e., spatial versus non-spatial)
has led to two conflicting views for the functional specialization of the lateral prefrontal cortex.
In the first view, the lateral prefrontal cortex is specialized according to the type of information
being processed (Levy and Goldman-Rakic 2000). That is, spatial information is mediated
through the DLPFC, while non-spatial information is mediated by the ventrolateral prefrontal
cortex consistent with the dorsal ―what‖ and ventral ―where‖ visual pathways (Curtis and
D'Esposito 2004). By contrast, the second view suggests a process-specific model for working
memory that emphasizes the importance of the level of processing required for the task rather
than the sensory modality (Petrides 1995; Petrides 2000). In this model, a hierarchy suggests that
the ventrolateral prefrontal cortex is involved in the active encoding and retrieval of information,
18
while the mid-dorsolateral prefrontal cortex supports higher order executive control functions
such as the monitoring and manipulation of stored information (Curtis and D'Esposito 2004).
Alternatively, working memory has been considered as a process within long-term working
memory rather than a separate mechanism. For example, Cowan‘s model argues against devoted
brain region or buffers specialized to mediate different information contained in working
memory; and instead, suggest that the information contained in working memory activates a
portion of long-term memory that is currently being attended to (Cowan 1997). Support for
Cowan‘s model is starting to emerge with evidence from neuroimaging studies. For example,
work by the D‘Esposito group (Druzgal and D'Esposito 2003; Ranganath, Cohen et al. 2004)
have demonstrated different activation depending on the type of stimuli being maintained in
working memory thereby supporting the notion that the neural systems that process the
information currently being focused on in working memory should activate the same systems
used to process initially or store information in long-term memory. Cowan‘s model of working
memory is similar to one put forth by Engle and colleagues (Kane and Engle 2000). In this
model of working memory, Engle‘s group emphasizes the centrality of goal maintenance and
interference control of working memory. This body of work suggests that an important aspect of
working memory is the ability to maintain goal representations that allow one to select task-
relevant information from task-irrelevant information in the protection of information that may
be distracting or interfering. In this view, working memory is regarded as controlled attention.
Moreover, this conceptualization of working memory explains the relationship between working
memory capacity and variation in performance on working memory tasks such as the antisaccade
and the Stroop which are considered goal maintenance tasks (Kane and Engle 2003; Unsworth,
Schrock et al. 2004).
19
1.6.1 Summary
Since Baddeley‘s first proposal, a number of different theories of working memory have
emerged. Such theories, however, conflict with regards to their emphasis on a unitary nature of
working memory (Engle, Cantor et al. 1992), whereas others emphasize the non-unitary nature of
working memory and argue for a more domain-specific view of working memory (e.g.
(Daneman and Tardif 1987)). In addition, others have emphasized the individual differences in
working memory capacity and are conceptualized in terms of variation in the total amount of
mental resources available (e.g. (Just and Carpenter 1992)); however, these theories are
conflicted by those that claim that long-term knowledge and skills may attribute to individual
differences in working memory capacity (e.g. (Ericsson and Kintsch 1995)). Despite the number
of theories regarding working memory, it is evident that many brain regions are involved in this
process with the DLPFC being the most important.
1.6.2 Assessment of Working Memory
1.6.2.1 Neuropsychological Assessment
Several neuropsychological tests are used to assess working memory and are typically
administered as part of a cognitive battery, such as the Wechsler Adult Intelligence Scale, and
the MATRICS battery. Neuropsychological tests assess working memory across sensory
domains including non-verbal (e.g., visual, spatial, visuospatial, audiospatial and executive) and
verbal (sometimes referred to as phonological). For example, the delayed match to sample task
can test visual, spatial, visuospatial, auditory, audiospatial working memory in which the
stimulus is briefly presented to the subject followed by a delay period. Another stimulus is then
presented and the subject is required to determine if the current stimulus was the same as the one
previously presented. An example of a spatial working memory task is one in which 10 cubes are
irregularly spaced on a board and the subject is required to tap the cubes in the same or reverse
20
order as the test administrator. Furthermore, the number of blocks that the subject is required to
remember is increased with each trial thereby increasing the cognitive demand or working
memory load. An example of an executive working memory task is the self-ordered pointing task
in which subjects are presented a number of images or words which are arranged on a display.
Several trials are subsequently presented each with a different arrangement that contain some of
the previous stimuli and subjects are required to point to stimuli that they had not pointed to in
the previous trials. By contrast, a test to assess verbal working memory is the number-letter-span
in which a list of both numbers and letters is read out and the subject is required to mentally
reorder the string of numbers and repeat them to the administrator. The number-letter-span also
increases in working memory load with each successful trial. Another common assessment of
working memory is the classic Sternberg test. In the Sternberg task, lists are presented with
varying numbers of letters or words followed by a delay or retention period. Following the
retention period, a probe then is presented and the subject is required to answer yes or no if the
probe was on the list. The final working memory test that will be reviewed in this section is the
N-back task. In this task, letters are presented one at a time and subjects are required to
determine if the same letter was presented ―N‖ trials back. Typically, this task is administered at
increasing working memory loads (e.g., 0-3 back); however, the validity of the 0 and the 3-back
has been debated. That is, in the 0 back condition there is no memory component and therefore
has been criticized for serving as a control condition for this task and may only reflect attentional
components. The 3-back task condition has also been debated for its validity due to the ability of
subjects to successfully perform this task typically resulting in an inverted u-shaped curve in
healthy subjects (Callicott, Mattay et al. 1999). For example, it has been suggested that the 3-
back may involve attentional components (Kane, Conway et al. 2007) and/or short term memory
aspects of working memory (Shelton, Metzger et al. 2007). The drawback of the N-back task is
21
that the different subprocesses of working memory such as encoding, maintenance, and retrieval
cannot be examined. Finally, the N-back task is different from the other working memory tasks
reviewed because it involves continuous updating from trial to trial.
1.6.2.1.1 Neuropsychological Evidence for Working Memory Deficits in Schizophrenia
Goldman-Rakic (1994) proposed a neuropsychological model of schizophrenia with working
memory impairment as the core feature. Moreover, Goldman-Rakic specifically related working
memory dysfunction to a formal thought disorder which she argued was characterized by a
failure to adequately retain recent information or ideas in working memory. There has been
however some debate as to whether the putative neuropsychological impairment in schizophrenia
is a generalized one (Blanchard and Neale 1994) or rather certain functions are differentially
impaired. That is, increased distractibility (Oltmanns, Ohayon et al. 1978), memory impairment
(McKenna, Tamlyn et al. 1990), attentional disturbance (Nuechterlein and Dawson 1984) and
executive function (Weinberger, Berman et al. 1986) have all been suggested as core features of
schizophrenia. It is difficult though to distinguish impaired neuropsychological functions from
one another as these tests tend to vary in their degree of difficulty and sensitivity to the effects of
impaired functioning (Chapman and Chapman 1973) in addition to differences in intellectual
abilities among patients with schizophrenia (Forbes, Carrick et al. 2009). In this regard, Silver et
al (2003) tested the hypothesis that impaired working memory is a core deficit underlying
multiple neuropsychological deficits in patients with schizophrenia (Silver, Feldman et al. 2003).
Specifically, the relationship between working memory and other neuropsychological functions
was examined in 27 chronic male patients with schizophrenia and revealed significant
correlations between verbal and spatial working memory and other neuropsychological tests
(Silver, Feldman et al. 2003). Such significant relationships were not observed in healthy
22
subjects therefore supporting the hypothesis that working memory is a core deficit in
schizophrenia.
A wealth of studies has examined working memory function in patients with schizophrenia
which has produced robust and reliable reports of deficits across a range of measures (Lee and
Park 2005). For example, a recent meta-analysis performed on 187 studies that examined
working memory in schizophrenia compared to healthy subjects revealed deficits in all domains
of working memory (Forbes, Carrick et al. 2009). First, studies that examined verbal working
memory tests such as the digit span, letter-number span, and passage recall showed that patients
with schizophrenia performed significantly worse compared to healthy subjects (Forbes, Carrick
et al. 2009). Second, analysis on tests of visuospatial working memory including the spatial span
forwards and backwards, tests of pattern recognition, and the complex figure reproductions tests
demonstrated significant impairment in patients with schizophrenia compared to healthy subjects
(Forbes, Carrick et al. 2009). Lastly, patients with schizophrenia were also found to perform
significantly worse than healthy subjects on executive working memory tests such as self-
ordered pointing task, executive golf task, and the random number/letter generation (Forbes,
Carrick et al. 2009). Importantly, no relationship was found between current IQ with task
performance therefore indicating that the robust finding of impaired working memory function in
patients with schizophrenia was not simply attributed deficits in IQ (Forbes, Carrick et al. 2009).
The advancement of neuroimaging techniques has allowed for further investigation on the nature
of working memory deficits in patients with schizophrenia and will be reviewed in the following
section.
23
1.6.2.2 Review of Neuroimaging Techniques
1.6.2.2.1 Positron Emission Tomography
PET is a three-dimensional functional imaging technique that was first introduced in the 1950s.
This techniques involves the labeling of radioactive chemicals injected into the bloodstream
(Ter-Pogossian, Phelps et al. 1975). The system then detects pairs of gamma rays emitted by the
radionuclide (tracer) that accumulates in different regions of the brain depending on the
cognitive task being performed in the scanner. That is, radioligands selectively bind to specific
neurotransmitter receptors that will allow the visual reconstruction of the brain areas involved
and the mechanism underlying the cognitive task being tested. However, one disadvantage of
PET is the potential of ionizing radiation exposure.
1.6.2.2.2 Functional MRI
Magnetic resonance imaging (MRI) was introduced in the 1970s and is a detailed imaging
technique to visualize internal structures of the body. Moreover, MRI uses a powerful magnetic
field that aligns the nuclear magnetization of the hydrogen atoms of water in the body. This
technique is particularly useful in imaging the brain as white and gray matter contains different
amount of water thereby generating different contrasts in the MRI. Seigi Ogawa later extended
this method and introduced functional MRI (fMRI) in the 1990s which allows the visualization
of activated neuronal tissue (Ogawa, Lee et al. 1990). Functional MRI is based on the presence
of deoxyhemoglobin in blood that changes the proton signal from water molecules surrounding a
blood vessel in gradient echo MRI which generates blood oxygenation level-dependent (BOLD)
contrast. Functional MRI therefore allows the visualization of BOLD activation in neuronal
tissue while human subjects perform cognitive tasks. Spatial resolution is the advantage of MRI
over other techniques such as electroencephalogram (EEG), while temporal resolution is limited.
24
Moreover, activation of the BOLD response does not indicate the nature of neuronal activity
(i.e., inhibitory versus excitatory).
1.6.2.2.3 Magnetoencephalography
David Cohen (1968) first introduced MEG which is a technique similar to EEG (method
described in the next section) that measures the magnetic fields induced by the electrical currents
that naturally occur in the brain. MEG is advantageous over EEG since the magnetic fields are
not distorted by the skull and requires less preparation. MEG, however, is more costly than EEG
as this technique requires a magnetically shielded room and the equipment must be kept cool by
liquid helium. Current MEG machines involve a helmet-shaped dewer that contains
approximately 300 sensors that is placed just above the subject‘s head. Another disadvantage of
MEG is that it takes approximately 50 000 active neurons to generate a signal and these signals
mostly originate from tangential dipoles located in the sulci wall compared to EEG that is more
sensitive to radial dipole sources (Srinivasan, Winter et al. 2006).
1.6.2.3 Neuroimaging Working Memory
A rich literature exists on the examination of working memory with neuroimaging techniques.
Such studies have attempted to support the multiple theories of working memory particularly
with regards to unitary versus non-unitary systems. First, monkey studies have shown through
single unit recordings that increased firing in the neurons of the lateral prefrontal cortex during
the delay period of working memory tasks (Fuster 1995; Goldman-Rakic 1995). Moreover,
monkey lesion and neurophysiological studies suggest that the DLPFC is the brain region
specialized for the storage of spatial information while the ventrolateral prefrontal cortex is
involved in the storage of non-spatial information. Such division of domains have been mapped
onto the conventional ‗what‘ (non-spatial) and ‗where‘ (spatial) visual pathways of the brain
25
(Curtis and D'Esposito 2004). In line with this suggestion, monkey anatomical studies have
demonstrated that the parietal region that is a specialized for vision predominantly projects onto
the DLPFC (Petrides and Pandya 1984; Cavada and Goldman-Rakic 1989), whereas the
temporal cortex that is specialized for object vision projects more to the ventrolatral prefrontal
cortex (Barbas 1988). Functional neuroimaging studies in humans, however, have failed to
demonstrate this anatomical division between spatial and non-spatial information (D'Esposito,
Aguirre et al. 1998). Instead, this analysis plotted all the functional imaging studies up to the
year of its publication and found a hemispheric organization for spatial and non-spatial
information rather than a DLPFC versus ventrolateral prefrontal cortex, respectively. That is, the
findings of D‘Esposito et al (1998) demonstrate that the left hemisphere is specialized for the
mediation of non-spatial information whereas the right hemisphere mediates spatial information
(D'Esposito, Aguirre et al. 1998).
Considering the wealth of neuroimaging studies on working memory with objectives to
investigate the multiple theories of this cognitive process, this section will limit its discussion to
studies that have employed the N-back task in healthy subjects as this task paradigm was
employed in the experiments described in the proceeding chapters. In this regard, Owen et al
(2005) conducted a meta-analysis on 24 studies that employed the N-back task testing both
manipulation (location versus identity monitoring) and content (verbal versus non-verbal)
variants in healthy subjects. Across studies, robust activations were found in the lateral and
medial premotor cortex, dorsal cingulate, DLPFC, ventrolateral prefrontal cortex, frontal poles,
and the medial and lateral posterior parietal cortex (Owen, McMillan et al. 2005). Subsidiary
meta-analysis on the primary data revealed similar broad activation patterns for identity
monitoring of verbal information and both location and identity monitoring of non-verbal
26
information. This analysis therefore reveals the involvement of a frontoparietal system in the
mediation of both verbal and non-verbal working memory tasks (Owen, McMillan et al. 2005).
1.6.2.3.1 Neuroimaging Evidence for Working Memory Deficits in Schizophrenia
The nature of abnormal brain activation while performing working memory tests in fMRI
remains controversial owing to the inconsistent findings in literature. That is, some studies have
found evidence of enhanced activation of the DLPFC (Callicott, Bertolino et al. 2000; Manoach,
Gollub et al. 2000), no difference (Honey, Bullmore et al. 2002; Walter, Wunderlich et al. 2003;
Kindermann, Brown et al. 2004; Walter, Vasic et al. 2007), and reduced activity (Callicott,
Ramsey et al. 1998; Barch, Carter et al. 2001; Perlstein, Carter et al. 2001; Barch, Csernansky et
al. 2002) while patients with schizophrenia perform working memory tasks. It has also been
suggested that reduced activation of the DLPFC is predicted by adjacent white matter
disturbances shown through diffusion tensor imaging (Schlosser, Nenadic et al. 2007). Meta-
analyses performed on working memory studies in patients with schizophrenia also are
inconsistent with regards to the nature of DLPFC activation. For example, a recent meta-analysis
of 12 studies that tested patients with schizophrenia on the N-back task revealed significantly
lower activation of the DLPFC (Glahn, Ragland et al. 2005). By contrast, a more inclusive meta-
analysis on 29 studies of working memory failed to show a difference in DLPFC activation in
patients with schizophrenia compared to healthy subjects (Van Snellenberg, Torres et al. 2006).
However, the meta-analysis conducted by van Snellenberg et al (2006) revealed a relationship
across studies between DLPFC differences between patient and control groups and patient-
control differences in task performance. That is, poorer performance in patients with
schizophrenia was predictive of lower DLPFC activation thereby suggesting that the findings of
reduced DLPFC activation may be attributed to greater performance deficits in patients (Van
27
Snellenberg, Torres et al. 2006). In line with this finding, Callicott et al (2003) examined DLPFC
activation while performing the N-back task and compared high versus low performing subjects
in both patient and control samples (Callicott, Mattay et al. 2003). When low-performing patients
were compared to either low or high performing healthy subjects, reduced DLPFC was found,
whereas, when high performing patients were compared to low performing healthy subjects,
increased DLPFC was revealed in patients with schizophrenia (Callicott, Mattay et al. 2003).
Furthermore, when high performing patients were compared to high performing healthy subjects,
discrete regions of the DLPFC were found to be more active thereby suggesting that patients
with schizophrenia may employ a different brain network compared to healthy subjects in order
to achieve a greater level of performance on the N-back task (Callicott, Mattay et al. 2003).
In addition to task performance, other factors such as medication or physical gray matter
differences have been suggested to contribute to the inconsistent activation of the DLPFC during
working memory performance. For example, several studies have reported that patients treated
with second generation antipsychotic medication show increased prefrontal activation during
working memory tasks (Honey, Bullmore et al. 1999; Meisenzahl, Scheuerecker et al. 2006;
Wolf, Janssen et al. 2007) however this effect was not related to working memory performance.
The impact of medication on DLPFC therefore remains unclear. Further, greater heterogeneity in
DLPFC activation in patients with schizophrenia due to differences in the morphology of the
prefrontal cortex may also result in decreased DLPFC when converted to a standard template
(Manoach, Gollub et al. 2000). Finally, differences in DLPFC may also result from the domain
of working memory tested (i.e., verbal versus non-verbal tests) and other task parameters
(Manoach 2003).
28
1.6.3 Working Memory: The Role of Neurotransmitter Systems
1.6.3.1 Dopamine System
Pharmacological manipulation in both animals and humans has suggested that dopamine may
enhance working memory performance. In monkeys, following 6-hydroxy-dopamine lesions in
the prefrontal cortex or the administration of dopamine antagonists impair working memory
function (Sawaguchi and Goldman-Rakic 1994). By contrast, the administration of low dose
dopamine agonists improves working memory performance in monkeys (Williams and
Goldman-Rakic 1995). Dopamine agonists have also been shown to improve working memory
function in humans. For example, the administration of non-selective agents, such as
amphetamine and methylphenidate, has been demonstrated to improve accuracy and reaction
time in healthy subjects (Barch and Carter 2005). Furthermore, Mintzer and Griffiths (2003)
observed improved performance on the 2-back condition of the N-back task (Mintzer and
Griffiths 2003), although it has been suggested that this effect may be limited to poor baseline
performers (Mattay, Berman et al. 1996). Finally, the administration of amphetamine has been
demonstrated to ameliorate deficits in working memory in sleep deprived individuals (Pigeau,
Naitoh et al. 1995; Magill, Waters et al. 2003). In general, the findings of these studies suggest
that amphetamine and methylphenidate may improve working memory performance though
there is some evidence that suggests that this effect may have a greater impact on those with
relatively poor baseline performance.
The role of dopamine in working memory has been suggested to increase the stability of active
representations in memory in order to adequately respond to task-relevant stimuli (Durstewitz,
Seamans et al. 2000). Braver and colleagues (1999) have also suggested that dopamine may
serve as a cue for updating information in working memory and that phasic dopamine signals
may help to regulate what information is encoded in working memory. Furthermore, Braver et al
29
suggest that dopamine's role in working memory may be to protect against interference of
distracting or competing information (Braver, Barch et al. 1999; Braver and Cohen 1999). Hazy
et al (2007) has further suggested that the basal ganglia mediated by dopamine signals serve to
update representations in working memory (Hazy, Frank et al. 2007).
1.6.3.1.1 Pharmacological Manipulation of Dopamine on Working Memory in Schizophrenia
There are a few studies which have examined the impact of dopamine on working memory
function in patients with schizophrenia. For example, the administration of apomorphine to
medication withdrawn patients with schizophrenia resulted in no significant change in
performance, but a significant task-related increase in blood flow to the DLPFC was observed
(Daniel, Berman et al. 1989). More recently, the administration of amphetamine in medicated
patients with schizophrenia has shown some promise. For example, Barch and Carter (2005)
administered amphetamine and a placebo-control to stable medicated patients with schizophrenia
which was found to enhance performance on a spatial delayed match to sample task evidenced
through improved accuracy and reaction time (Barch and Carter 2005). However, improved
performance was observed in conditions with and without a delay period thereby suggesting that
amphetamine increased the encoding subprocess of the working memory task rather than the
storage of information (Barch and Carter 2005). These studies therefore provide reasonable
evidence for the positive impact of amphetamine on working memory performance in medicated
patients with schizophrenia through the augmentation of dopaminergic function.
1.6.3.2 Noradrenergic System
The administration of noradrenergic agonists have also been demonstrated to have a positive
influence on working memory particularly in animals while less consistent findings have been
found in humans. For example, a couple of studies have administered noradrenergic alpha-2
30
agonists in monkeys which resulted in improved working memory performance, particularly in
aged animals (Arnsten, Cai et al. 1995; Arnsten and Goldman-Rakic 1998). Such improvement
in working memory function with alpha-2 agonists was suggested to protect against interference
thereby decreasing distractibility (Arnsten and Contant 1992). In humans, however, there has
been mixed results in terms of the effect of alpha-2 agonists on working memory function. For
example, Coull et al (1995) found that alpha-2 agonist clonidine improved spatial self-ordered
pointing in healthy subjects but this effect was only observed if subjects were well-practiced and
with 2.5 µg/kg, while administration of 1.5 µg/kg doses impaired performance (Coull, Middleton
et al. 1995). Other studies by Jakala et al (1999a, 1999b) demonstrated that clonidine at both low
(0.5, 2 µg/kg) and higher doses (5 µg/kg) impaired working memory performance on the spatial
delayed match to sample and the self-ordered pointing tasks evident through an increased
number of errors or by the increase in reaction time (Jakala, Riekkinen et al. 1999; Jakala, Sirvio
et al. 1999). The current literature on the impact of alpha-2 agonists on working memory
performance is positive in non-human primates however this effect has been inconsistent with
human subjects. If patients with schizophrenia have impaired noradrenergic function then it is
possible that alpha-2 agonists may have a positive impact on working memory performance in
this disorder.
1.6.3.3 Acetylcholine System
The influence of the cholinergic agents on working memory has also been examined. For
example, Furey et al (1997) demonstrated that the cholinesterase inhibitor physostigmine
improved performance on the face delayed to match sample task (Furey et al 1997) that may be
related to a reduction of task-related prefrontal activity and increased activity in the extrastriate
brain regions (Furey, Pietrini et al. 2000; Furey, Pietrini et al. 2000). The authors suggested that
increased activity in the visual cortex may have improved the encoding of visual information,
31
which could have contributed to improved working memory performance (Furey, Pietrini et al.
2000). The studies conducted by Furey and colleagues suggest that cholinergic agents may have
a positive impact on working memory in patients with schizophrenia but more studies are
needed.
1.6.3.4 Serotonin System
The literature on the role of the serotonin system in working memory is limited. Luciana et al
(1998) administered a serotonin agonist fenfluramine to human subjects and observed a decrease
in delayed spatial memory (Luciana, Collins et al. 1998). A later study by this group
administered tryptophan which is a precursor to serotonin and demonstrated impairment in
performance on the digit span backwards and affective working memory (Luciana, Burgund et
al. 2001). Together these two studies suggest that enhancing serotonin activity would not have a
positive benefit on working memory in human subjects.
1.6.3.5 GABAergic System
Although working memory is considered an emergent property of a neuronal network that is
involves a number of different brain regions (Goldman-Rakic 1988), it is reliant on the
coordinated and sustained firing of neurons within the DLPFC to mediate the temporary
presentation of a stimulus cue and the initiation of the behavioural response in working memory
(Goldman-Rakic 1995). As reviewed in a previous section, DLPFC neurons use GABA as the
principle neurotransmitter and thus the role of GABA in working memory has been extensively
studied.
Inhibitory interneurons that use GABA are believed to provide the mechanism in which to
synchronize pyramidal neuronal activity during working memory processes (Lewis, Hashimoto
et al. 2005). This proposal has been supported by several studies in monkeys. For example,
32
during the delay period of working memory tasks, Wilson et al (1994) demonstrated sustained
activity in the DLPFC of monkeys (Wilson, O'Scalaidhe et al. 1994) that has been suggested to
be important in both the task-related neuronal firing and the spatial tuning of neuronal responses
during working memory (Rao, Williams et al. 2000). In line with this suggestion, Sawaguchi et
al (1989) injected GABA antagonist bicuculline into the DLPFC of monkeys and observed a
disruption in working memory performance (Sawaguchi, Matsumura et al. 1989). However,
decrements in working memory performance following the administration of baclofen, a GABAB
receptor agonist has been shown in rats (Nakagawa and Takashima 1997; Romanides, Duffy et
al. 1999). By contrast, in humans baclofen was shown to enhance working memory performance
(Escher and Mittleman 2004).
1.6.3.5.1 Pharmacological Manipulation of GABA on Working Memory in Schizophrenia
Pharmacological manipulations of GABAA receptors have been examined for its effect on
working memory in patients with schizophrenia compared to healthy subjects (Menzies, Ooi et
al. 2007). In this study, subjects received either GABAA agonist (lorazepam), GABAA antagonist
(flumazenil) or a placebo prior to the performing the N-back task (Menzies, Ooi et al. 2007).
This study found that lorazepam impaired performance while flumazenil enhanced it and this
effect was most pronounced in patients with schizophrenia compared to healthy subjects. Taken
together, these studies suggest that inhibitory activity of the DLPFC mediated through GABA
has a spatial tuning and temporal role that is critical to working memory (Constantinidis,
Williams et al. 2002) and that pharmacological manipulations of GABA may improve
performance in patients with schizophrenia.
33
1.6.3.5.2 Cortical Inhibition
Cortical inhibition refers to a neurophysiological mechanism in which inhibitory neurons
selectively attenuate the activity of other neurons in the cortex. Pyramidal cells comprise
approximately 70-80% of the cortex while interneurons make up about 15-30% (DeFelipe and
Farinas 1992). Interneurons are predominantly mediated by the inhibitory neurotransmitter
GABA and are classified according to the type of synapse they form (Benes and Berretta 2001).
The most common cortical interneurons are basket cells which form axo-somatic inhibitory
synapses with pyramidal cells around the surface of the pyramidal cell body (Benes and Berretta
2001). Moreover, basket cells are located in layers III-IV and have long axons that ascend to
more superficial layers of the cortex and are also modulated by subcortical activity through direct
thalamic inputs (Benes and Berretta 2001). Chandelier cells make up the second type of
interneurons that form axo-axonal inhibitory synapses with the initial segment of pyramidal cells
axons (Somogyi, Tamas et al. 1998; Lewis, Hashimoto et al. 2005). This configuration has been
suggested to allow chandelier inhibitory interneurons to have a strong influence over pyramidal
cell output (Lewis, Hashimoto et al. 2005) possibly through the modulation of the responsity of
other cortical neurons (Benes and Berretta 2001). Double bouquet cells make up the third
classification of interneurons that form an axo-dendritic inhibitory synapse on the apical and
basal dendrites of pyramidal cells in addition to adjacent interneurons (Somogyi and Cowey
1981). Further, double bouquet interneurons may be involved in the disinhibition of pyramidal
cells to attenuate excessive inhibitory synaptic control (Somogyi, Tamas et al. 1998; Benes and
Berretta 2001). Basket and Chandelier interneurons are located in cortical layers III-IV and are
fast-spiking, while double bouquet interneurons are found in cortical layer II and III (Benes and
Berretta 2001).
34
Cortical excitation reflects the balance between excitatory and inhibitory activity. Moreover,
pyramidal cell activity results from both excitatory post-synaptic potentials (EPSPs) and
inhibitory post-synaptic potentials (IPSPs) that terminate throughout the cell (Krnjevic 1997).
This coordinated interaction of EPSPs and IPSPs is universally accepted as an important element
in the control of neuronal firing. Cortical inhibition therefore reflects the activity of IPSPs
generated by GABAergic interneurons and plays several important functions in the cortex
including working memory (Lewis, Pierri et al. 1999; Constantinidis, Williams et al. 2002).
Specific subclasses of GABA inhibitory interneurons in the DLPFC may play a critical role in
regulating the activity of layer III pyramidal neurons that have been demonstrated to be
important in working memory (Fuster, Bauer et al. 1985; Friedman and Goldman-Rakic 1994).
For example, the spatial arrangement of the axon terminals of basket cells in which a large
portion of their terminals synapse on the soma of pyramidal cells may provide a reasonable
mechanism to inhibit the activity of pyramidal neurons located in the gaps between the
interconnected stripes (Pucak, Levitt et al. 1996; Melchitzky, Sesack et al. 1998). Chandelier
cells may also provide inhibition of layer III pyramidal neurons through the formation of vertical
arrays that provide inhibitory input exclusively to the axon initial segment of pyramidal cells
(Peters 1984). In addition, the spread of chandelier cell‘s axon arbors is consistent with the width
of the stripes formed by later III pyramidal cells further suggesting that these cells may be
important in the regulation of pyramidal cell output (Lund and Lewis 1993). In the following
section, the role of GABA in cortical oscillations will be discussed followed by the evidence for
oscillatory activity underlying working memory function.
35
1.6.3.6 Neurophysiological Assessment of Working Memory
1.6.3.6.1 Electroencephalography
In the 1920s Hans Berger recorded brain waves from the human scalp and named this technique
EEG. EEG is one of the most widely used techniques in which to record brain activity due to its
high temporal resolution and low cost. EEG involves the placement of electrodes on the scalp
and therefore mostly captures the synaptic activity from post synaptic potentials of neurons in the
superficial layers of the brain (Buzsaki, 2006). Since the electric potential from a single neuron is
too small to be captured by EEG, the activity recorded reflects the summation of the synchronous
activity from thousands of neurons with a similar spatial orientation. Furthermore, during
neuronal processing, EEG captures a finite number of discrete frequency bands that are
conventionally divided into delta (1-4 Hz), theta (4-7 Hz), alpha (8-12 Hz), beta (12-28 Hz), and
gamma (30-50 Hz) oscillatory activity (Buzsaki 2006).
1.6.3.7 Cortical Oscillations
Over the past two decades, there has been a shift from examining event related potentials to brain
oscillations following the discovery that cortical networks have a natural tendency to engage in
oscillatory activity at multiple frequencies (Singer 1999; Buzsaki and Draguhn 2004; Fries,
Nikolic et al. 2007). Oscillations measured by both in vitro (i.e., brain slices and patch-clamp
recording) and in vivo through scalp recordings with EEG and MEG have been associated
activity within different frequency bands to different states of the brain. For example, lower
oscillatory activities including delta, theta, and alpha are associated with different stages of sleep,
while higher frequencies have been associated with the waking brain (Buzsaki 2006).
The mechanisms underlying the generation of cortical oscillations are currently being
investigated. At the single neuronal level, the gating of ion channels in the cell membrane has
36
been suggested to provide neurons with intrinsic properties that allow them to oscillate at
different frequencies (Buzsaki 2006). That is, the open or closed state of these membrane
channels is under the control of the membrane voltage, neurotransmitters and modulators, in
addition to other cell factors that determine how eagerly and precisely the neuron responds to a
given input. In addition to voltage gated channels, ligand-, ion-, and second messenger gated
channels endow neurons with the intrinsic properties capable of generating a repertoire of
activities including oscillation and resonance at multiple frequency ranges (Buzsaki 2006).
Furthermore, several different ion compositions exist across various neuronal cell types which
allows for cortical neurons to have a wide range of preferred frequencies, spike timing
properties, and their diverse frequency-tuning properties are critical for setting network dynamics
(Buzsaki 2006). For example, GABAergic interneurons respond with the highest temporal
precision to frequencies in the gamma band, while pyramidal neurons respond more precisely to
activities in the lower frequency ranges (Gupta, Wang et al. 2000; Thomson 2000; Markram,
Toledo-Rodriguez et al. 2004).
1.6.3.7.1 Cortical Oscillations and Information Encoding
Interesting relationships have been uncovered between frequency bands that may be important in
information encoding and functional roles in the cortex. For example, the power of a cortical
oscillation has been shown to decrease as the frequency of the oscillation increases. Such a
relationship suggests that a temporal correlation exists between frequency bands therefore
oscillatory activity at higher frequencies are vulnerable to alterations at lower frequency bands
(Buzsaki 2006). Furthermore, an inverse relationship has been shown between the size of the
neuronal pool and the frequency of oscillatory activity. Thus, a small number of neurons tend to
participate in higher frequency oscillatory activity whereas a large number of neurons are needed
to participate in slower frequencies (Buzsaki 2006). In addition, activities between frequency
37
bands have been shown to interact with each other which may be important in information
processing. For example, frequency concatenation occurs when two local networks that oscillate
at distinct frequency bands are co-activated and interact to generate activity in a new frequency
with a period that is the sum of the two networks (Roopun, Kramer et al. 2008). Another
example is cross-frequency interaction where the power of discrete frequency band is modified
by the phase of a lower frequency band that co-exists during information processing (Roopun,
Kramer et al. 2008).
Information processing has also been linked to the modulation of cortical oscillations (Fries,
Nikolic et al. 2007). That is, Fries et al (2007) suggested that the modulation of oscillatory
responses of neurons may provide the temporal frame that determines which neurons
communicate with each other (Fries, Nikolic et al. 2007). The resulting neuronal responses have
further been suggested to convey information in two different orthogonal messages in parallel
(Singer 1999). First, neurons should indicate the presence of the feature to which they are tuned;
and second, neurons should indicate with which other neurons they are communicating with
(Singer 1999). The first message is encoded in the discharge frequency of the neurons and it is
suggested that the second message is contained in the precise timing relationships between
individual spikes of distributed neurons thus providing the temporal code (Singer 1999). Singer
further proposed that the precise timing relations are achieved either by internal timing
mechanisms or by the timing of external events (stimulus locking). Internal timing mechanisms
involve an oscillatory modulation of neuronal responses across different frequency bands
ranging from delta to gamma ripples. These oscillations therefore limit the communication of
cells to short temporal windows and the duration of these windows decreases with oscillation
frequency (Singer 1999). Through the varying phase relationship between oscillating groups,
networks of functionally cooperating neurons can be flexibly configured within hard wired
38
networks. Furthermore, the synchronization of spike emitted by neuronal population achieves
greater saliency of their responses and is enhanced due to the coincidence sensitivity of receiving
neurons by increasing the discharge rate (Singer 1999). Such a temporal code through the
modulation of oscillatory activity may underlie the association between working memory and
oscillatory activity.
1.6.3.7.2 Cortical Oscillations Provides the Temporal Framework for Working Memory
Temporal segmentation by phase encoding has been suggested to be involved in the maintenance
of multiple working memory items (Jensen 2006). For example, if 5 elements A through to E are
to be maintained in a working memory buffer through the repeated rehearsal of the list A-E. This
mechanism involves a fast rate at which the individual items are activated and a slow rate at
which the full list is repeated. Although this scheme was first proposed as a mechanism
underlying short term memory (Horn and Usher 1992), Lisman and Idiart (1995) suggested that
the two activation rates are associated with brain oscillations particularly in the theta and gamma
frequency ranges. Furthermore, Lisman and Idiart suggested that the number of gamma cycles
per theta cycle determines the working memory capacity of the buffer and that the gamma
frequency determines how fast items can be retrieved (Lisman and Idiart 1995). The principle of
dual oscillations can therefore account for how multiple items are maintained and retrieved in
working memory (Jensen 2006).
1.6.3.8 Cortical Oscillations Association with Working Memory
Evoked and induced oscillatory activity has been associated with performing both verbal and
non-verbal working memory tasks. Several studies have shown activations in the theta
oscillatory activity during working memory tasks in healthy subjects (Sarnthein, Petsche et al.
1998; Klimesch, Doppelmayr et al. 2001; Raghavachari, Kahana et al. 2001; Jensen and Tesche
39
2002; Schack, Vath et al. 2002). Fewer studies report activations in the alpha (Gevins, Smith et
al. 1997; Klimesch, Doppelmayr et al. 1997; Jensen, Gelfand et al. 2002; Rizzuto, Madsen et al.
2003) and beta (Tallon-Baudry and Bertrand 1999; von Stein, Rappelsberger et al. 1999; Schack,
Vath et al. 2002) frequency bands while healthy subjects perform working memory tasks.
Finally, activations in the gamma frequency range have been consistently associated with
working memory in healthy subjects (Sarnthein, Petsche et al. 1998; Tallon-Baudry and Bertrand
1999; Tallon-Baudry, Bertrand et al. 2001; Gruber and Muller 2005; Kaiser and Lutzenberger
2005).
1.6.3.8.1 Gamma Oscillations and Working Memory
Activations in the gamma frequency band have been correlated with specific working memory
processes. For example, Howard et al (2003) tested epileptic patients who were undergoing
intracranial EEG recordings prior to surgical resection in the classic Sternberg task and observed
increases in gamma power in response to increases in working memory load (Howard, Rizzuto et
al. 2003). Such a relationship between gamma power and working memory load has also been
demonstrated in healthy subjects using both EEG and MEG (Tallon-Baudry, Bertrand et al.
1998; Mainy, Kahane et al. 2007; Meltzer, Zaveri et al. 2008). Increases in gamma power in
response to working memory load have also been reported during non-verbal working memory
tasks (Lutzenberger, Pulvermuller et al. 1995; Tallon-Baudry, Bertrand et al. 1998;
Lutzenberger, Ripper et al. 2002; Pesaran, Pezaris et al. 2002). Gamma oscillatory activity has
been suggested to organize and temporally segment the representations of both verbal and non-
verbal items in a multi-unit working memory system (Lisman and Idiart 1995).
Gamma band activity during working memory tasks is observed in distributed network (Mainy,
Kahane et al. 2007) and includes focal centres in the prefrontal cortex (BA 9, 10, 44, 45, 46), the
40
precentral and postcentral gyri, the auditory cortex, the fusiform gyrus, and the hippocampus
which is consistent with human lesion (Petrides and Milner 1982; Frisk and Milner 1990; Owen
and Craddock 1996), and neuroimaging (Jonides, Smith et al. 1993; Braver, Cohen et al. 1997;
Cohen, Perlstein et al. 1997; Courtney, Ungerleider et al. 1997; Courtney, Petit et al. 1998;
Martinkauppi, Rama et al. 2000; Romanski 2004) studies. Furthermore, neuroimaging studies
have demonstrated that the BOLD signal in functional MRI is frequency dependent. That is,
BOLD responses were found to be positively correlated at higher frequencies including the
gamma frequency range, while activity within the lower frequency ranges (i.e., alpha power) was
correlated in a negative fashion (Logothetis, Pauls et al. 2001; Foucher, Otzenberger et al. 2003;
Moosmann, Ritter et al. 2003; Brookes, Gibson et al. 2005; Mukamel, Gelbard et al. 2005;
Niessing, Ebisch et al. 2005). These studies together suggest that activations within the gamma
frequency band are involved in specific working memory subprocesses and are robustly observed
in the DLPFC that is critical to this cognitive process.
1.6.3.8.2 The Generation and Modulation of Gamma Oscillations
GABAergic receptor mediated inhibitory neurotransmission plays a critical role in generation
and modulation of gamma oscillations. There are several studies which suggest that GABAA
receptor mediated IPSPs contribute to the generation of gamma oscillations (Whittington, Traub
et al. 1995; Wang and Buzsaki 1996; Bartos, Vida et al. 2007). For example, GABAA receptors
typically discharge at a rate of 30-50 Hz (i.e., gamma frequency band) which induces a high-
frequency on-off oscillatory pattern of pyramidal cell discharge when recorded by EEG. In the
rat, Whittington et al (1995) demonstrated that the activation of glutamate receptors in
hippocampal and neocortical slices resulted in synchronous 40 Hz oscillatory activity in the
networks of inhibitory neurons that are connected by synapses comprising GABAA receptors.
Whittington et al (1995) further demonstrated that the administration of a GABAA antagonist
41
bicuculline resulted in the amelioration of these gamma oscillations (Whittington, Traub et al.
1995). In addition, Wang and Buzsaki showed through computer simulations that the synaptic
time constant for GABAA receptors approximately ranges from 10 to 25 milliseconds further
implicating GABAA receptor activity in the generation of gamma oscillations (Wang and
Buzsaki 1996). In line with this finding, the prolongation of the decay period of the GABAA
receptor post synaptic current with barbiturate was shown to reduce gamma oscillations (Fisahn,
Pike et al. 1998).
The modulation of gamma oscillations have been suggested to be achieved through GABAB
receptor mediated IPSPs (Whittington, Traub et al. 1995; Brown, Davies et al. 2007; Leung and
Shen 2007). In this regard, in vivo and in vitro studies have demonstrated that the activation of
GABAB receptors results in the suppression of both spontaneous and stimulus induced gamma
oscillatory activity. For example, the administration of baclofen, a GABAB agonist was shown to
ameliorate gamma oscillations in rat hippocampal slices (Brown, Davies et al. 2007), while the
blockade of GABAB receptors in the hippocampus of behaving rats results in an increase in
gamma oscillations (Leung and Shen 2007). Second, the reduction of parvalbumin has been
shown to result in increased facilitation of GABA release during repetitive synaptic firing has
been directly linked gamma modulation (Vreugdenhil, Jefferys et al. 2003; Sohal, Zhang et al.
2009). Finally, through transcranial magnetic stimulation (TMS) studies the activity of GABAB
receptors has been shown to have an inhibitory effect on GABAA receptor activity (Sanger, Garg
et al. 2001; Daskalakis, Christensen et al. 2002) with an inhibitory effect that typically ranges
from 250 to 500 milliseconds (McCormick 1989).
42
1.6.3.9 Analyzing Cortical Oscillations
Oscillatory activity recorded during working memory can be quantified by two different EEG
analytical techniques. The first method examines evoked oscillatory responses that are phase-
locked to the stimulus onset with a fixed latency within the first 100 msec following stimulus
onset and are measured by stimulus-triggered averaging of responses (Tallon-Baudry, Kreiter et
al. 1999). The second method involves the examination of induced oscillatory activity is not
phase-locked to stimulus onset and jitters in latency varying from trial to trial; thus, responses are
cancelled out when averaged (Tallon-Baudry, Kreiter et al. 1999). Although both evoked and
induced oscillatory activities have been associated with working memory (Howard, Rizzuto et al.
2003; Basar-Eroglu, Brand et al. 2007), induced oscillatory activity in the gamma frequency
band has been suggested to reflect microsaccadic eye movement rather than neuronal processing
during cognitive paradigms (Yuval-Greenberg, Tomer et al. 2008). That is, when Yuval-
Greenberg et al (2008) recorded EEG simultaneously with high resolution binocular eye tracking
demonstrated that increased gamma power reflected in the single-trial EEG response is in fact
transient in nature rather than an index of sustained neuronal oscillatory activity. Second, the
observed transient increase in gamma power was time locked to- and coincided with the onset of
miniature saccades, while the time course reflected the dynamics of these saccadic eye
movements. These findings therefore suggest that induced gamma band activity is a direct
consequence of miniature saccades rather than neuronal oscillations in the gamma frequency
range. Yuval-Greenberg et al (2008), therefore, contend that the evaluation of evoked rather than
induced gamma oscillatory activity mitigates the effect of miniature saccades on this
neurophysiological phenomenon (Yuval-Greenberg, Tomer et al. 2008). Additionally, gamma
band activity during cognitive tasks has been shown to reflect cranial musculature artifact
(Shackman 2010), however, since this activity is characterized by irregular spikes and waves
43
present in all spectral frequencies such artifact should be significantly reduced when multiple
trials are averaged with evoked analytical methods (Barr and Daskalakis 2010).
1.6.4 Neurophysiological Evidence for Working Memory Deficits in Schizophrenia
Oscillatory activity during working memory performance has started to be examined in patients
with schizophrenia. Recent studies have observed alterations in oscillatory activities across
spectral frequencies in patients with schizophrenia compared to healthy subjects.
1.6.4.1 Delta
Spectro-temporo-spatial MEG patterns have been examined recently by Ince et al (2009) while
patients with schizophrenia medicated with non-conventional antipsychotics and healthy subjects
while performing the Sternberg task. With this technique, changes in oscillatory amplitude are
captured. That is, when an event causes an amplitude decrease in rhythmic activity this is
referred to as event-related desynchronization (ERD), while increases in amplitude are referred
to as event-related synchronization (ERS) (Pfurtscheller and Lopes da Silva 1999; Neuper and
Pfurtscheller 2001). Furthermore, these oscillatory power patterns have been shown to reflect
increased cortical excitability with ERD and decreased or inhibited cortical excitability with ERS
(Neuper and Pfurtscheller 2001). This study revealed alterations in delta ERD/ERS patterns in
the dorso-frontal, occipital and left fronto-temporal brain regions in patients with schizophrenia
during the encoding stage of working memory (Ince, Pellizzer et al. 2009). Stephane et al (2008)
also have examined ERD/ERS patterns during verbal working memory in 10 patients with
schizophrenia on nonconventional antipsychotic medication compared to 11 healthy subjects.
Using the Sternberg task, reduced ERD/ERS patterns were found in patients with schizophrenia
compared to healthy subjects during the encoding and maintenance stages of working memory.
44
Furthermore, patients with schizophrenia showed a lack of delta activity in the left frontal and
left parietal brain regions compared to healthy subjects (Stephane, Ince et al. 2008).
1.6.4.2 Theta
In 2005, Schiemdt et al examined event related (induced) theta oscillatory during the N-back task
combined with a task switching task in 10 medicated in-patients diagnosed with schizophrenia
compared to 10 healthy subjects (Schmiedt, Brand et al. 2005). The N-back task was
administered at 3 different working memory loads: 0-2 back with the added component of
response instructions (i.e., which hand to respond with left or right) while EEG was recorded.
Schmiedt et al (2005) reported increased theta oscillatory activity with increased working
memory load and rule-switching in the fronto-central region in healthy subjects. Moreover, such
modulation of theta oscillatory activity with working memory and task-related increases was not
observed in any brain region in patients with schizophrenia (Schmiedt, Brand et al. 2005).
Alterations in theta have also been reported by Haenschel et al (2009) who examined both
induced and evoked oscillatory activities in 14 early-onset patients with schizophrenia compared
to healthy subjects (Haenschel, Bittner et al. 2009). In their study, a delayed discrimination task
was used to examine oscillatory activity during the encoding, maintenance and retrieval stages of
visual working memory at 3 different loads. Haenschel et al (2009) observed decreased evoked
theta oscillatory activity during the encoding stage and a decrease in induced theta activity
during the retrieval stage in patients with schizophrenia compared to healthy subjects
(Haenschel, Bittner et al. 2009).
1.6.4.3 Alpha
Bachman et al (2008) were the first to examine working memory in 29 pairs of discordant twins
with schizophrenia (Bachman, Kim et al. 2008). In this study, spatial working memory was
45
tested at 4 different working memory loads to examine ERD/ERS patterns with MEG. Similar
ERD/ERS patterns over the posterior brain region at low working memory loads was observed;
however with increased working memory loads, patients and to an intermediate degree their non-
schizophrenic co-twins (both monozygotic and dizygotic pairs collapsed together) demonstrated
significantly greater increases in ERD/ERS patterns in the alpha frequency band compared to
healthy subjects (Bachman, Kim et al. 2008). Alterations in alpha oscillatory activity have also
been reported in the previously described studies by Ince et al (2009) and Stephane et al (2008).
By contrast, no differences were reported by Haenschel et al (2009) in induced alpha oscillatory
activity during the maintenance stage in patients with schizophrenia compared to healthy
subjects.
1.6.4.4 Beta
There are few reports of altered beta oscillatory activity in patients with schizophrenia during
working memory. In the study described previously, Stephane et al (2008) also reported an
absence of ERD/ERS beta activation in the left frontal and parietal lobes in patients with
schizophrenia compared to healthy subjects while performing the Sternberg task (Stephane, Ince
et al. 2008). In addition, decreased evoked beta oscillatory activity was observed in patients with
schizophrenia during the encoding stage of visual working memory compared to healthy subjects
(Haenschel, Bittner et al. 2009).
1.6.4.5 Gamma
Examination of gamma oscillatory activity in patients with schizophrenia during working
memory has yielded inconsistent findings. For example, studies have reported both increased and
decreased and no differences in gamma oscillatory activity in patients with schizophrenia
compared to healthy subjects. For example, Basar-Erolgu (2007) administered the N-back at 3
46
working memory loads (0-back, 1-back and 2-back) in 10 medicated in-patients with
schizophrenia compared to 10 healthy subjects and reported an increase in evoked gamma
oscillatory activity with working memory load in healthy subjects, while patients exhibited
increased gamma activity at each working memory load in the frontal brain region (Basar-
Eroglu, Brand et al. 2007). That is, regardless of task difficulty, patients elicited increased
gamma oscillatory activity even during the 0-back without any memory component. By contrast,
Lewis and colleagues (Cho, Konecky et al. 2006; Lewis, Cho et al. 2008) reported decreased
induced gamma oscillatory activity in the frontal region of 16 medicated patients with
schizophrenia compared to healthy subjects. In addition, these studies reported a lack of gamma
modulation with increased working memory load consistent with the study by Basar-Eroglu
(2007). More recently, Haenschel et al (2009) found no differences in either induced or evoked
gamma oscillatory during the maintenance stage of visual working memory. However, during the
retrieval stage, reduced induced gamma oscillatory activity was observed in patients with
schizophrenia compared to healthy subjects (Haenschel, Bittner et al. 2009).
1.6.4.6 Relationship with Behavioural Performance and Clinical Symptoms
Although the studies reviewed above demonstrate alterations in both the generation and
modulation of oscillatory activity while performing working memory tasks, these studies failed
to find a relationship between behavioural performance and oscillatory activity. Further, no
relationship has been uncovered between oscillatory activity and the severity of clinical
symptoms. Relationships have been uncovered between gamma oscillatory activity with both
positive and negative symptoms although through an auditory odd-ball task and not working
memory (Lee, Williams et al. 2003). In this study, patients were divided by symptom
47
predominance and found that positive symptoms correlated with increased gamma, while
negative symptoms were related to a reduction of gamma (Lee, Williams et al. 2003).
1.6.4.7 Summary
The lack of gamma modulation with working memory load in schizophrenia patients may be
attributed to alterations in GABA interneurons in the DLPFC (Akbarian, Kim et al. 1995; Benes
and Berretta 2001; Lewis, Hashimoto et al. 2005; Hashimoto, Bazmi et al. 2008) critical to
gamma oscillatory activity. Considering the importance of GABA in the generation
(Whittington, Traub et al. 1995; Wang and Buzsaki 1996; Bartos, Vida et al. 2007) and
modulation (Whittington, Traub et al. 1995; Brown, Davies et al. 2007) of gamma oscillations,
GABAergic impairments in patients with schizophrenia may underlie altered modulation of
gamma oscillatory activity while performing working memory tasks.
1.6.5 Treatment for Working Memory Deficits in Schizophrenia
1.6.5.1 Non-Pharmacological
Non-pharmacological treatments such as cognitive remediation programs that include drill and
practice exercises, teaching strategies to improve cognitive functioning, as well as, compensatory
strategies and group discussions have shown modest improvements on cognition, but with no
effects on working memory (McGurk, Twamley et al. 2007; Dickinson, Tenhula et al. 2010).
Other computer-based programs that focused on the remediation of verbal working memory in
schizophrenia through auditory training exercises have also shown promise (Adcock, Dale et al.
2009; Fisher, Holland et al. 2009). For instance, Fisher et al (2009) demonstrated significant
improvements on the letter number span working memory task in patients with schizophrenia
following 50 hours of auditory training exercises compared to a control group. Furthermore, this
48
training also improved auditory psychophysical performance that was found to be related to
improved verbal working memory and global cognition (Fisher, Holland et al. 2009).
1.6.5.2 Pharmacological
Pharmacological studies have also examined differences in antipsychotic medications on
cognitive functioning in patients with schizophrenia. While showing small effects toward
improved cognitive performance with treatment, some studies show therapeutic advantages of
second generation antipsychotics compared to conventional antipsychotics (Woodward, Purdon
et al. 2005) while others have not (Keefe, Bilder et al. 2007). Even clozapine, which is the
prototypic second generation agent for treatment resistant schizophrenia, was not shown to be
superior to other second generation antipsychotics suggesting that medications only result in
marginal therapeutic effects in schizophrenia (Harvey, Sacchetti et al. 2008). Several
pharmacological adjunct treatments have been shown to be of some efficacy for cognitive
dysfunction in schizophrenia. For example, the administration of galantamine, a combined
acetylcholinesterase inhibitor and allosteric potentiator of the nicotinic receptor, led to modest
improvements in attention (Bora, Veznedaroglu et al. 2005), delayed memory (Schubert, Young
et al. 2006), recognition (Lee, Lee et al. 2007), as well as colour naming on the Stroop test (Lee,
Lee et al. 2007) without any improvements in working memory. Barch and Carter (2005)
conducted a double-blind placebo-controlled study that examined the effects of amphetamine (D-
AMPH), a dopaminergic agonist, on cognition in schizophrenia and reported a decrease in
reaction time on spatial working memory and the Stroop task in addition to improved accuracy
across delay periods on the spatial working memory task (Barch and Carter 2005). Finally,
recombinant human erythropoietin is a neuroprotective agent that has also been examined for its
effects on cognition in chronic schizophrenia patients. In a double-blind, placebo-controlled
study, significant improvements on the Repeatable Battery for the Assessment of
49
Neuropsychological Status language-semantic fluency, attention-coding, delayed memory recall,
and working memory subtest attention-digit span compared to baseline measures following 12
weeks of recombinant human erythropoietin administration. These studies demonstrate that a
number of behavioural and pharmacological approaches have been investigated as possible
treatments for cognitive dysfunction in patients with schizophrenia. Although these studies show
some promise, it is clear that more efficacious treatments for cognitive deficits in schizophrenia
are still needed.
1.7 Repetitive Transcranial Magnetic Stimulation
Since Barker et al (1985) first introduced TMS as a non-invasive tool for the investigation of
motor cortex, repetitive applications of this technique has since been employed to study the
influence on a variety of cerebral functions. TMS involves the placement of an electromagnetic
coil over the scalp to produce an intense and localized magnetic field which can result in either
excitatory or inhibitory activation. Repetitive TMS (rTMS) uses alternating magnetic fields
applied at the same frequency to induce electric currents in the cortical tissue (Burt, Lisanby et
al. 2002). Low-frequency (≤ 1 Hz) rTMS is believed to cause inhibition of neuronal firing in a
localized area and is used to induce virtual lesions to examine a brain region‘s role in different
tasks. High-frequency rTMS (>1 Hz) is believed to be excitatory in nature and can result in
neuronal depolarization under the stimulating coil (Haraldsson, Ferrarelli et al. 2004). Moreover,
the effect of rTMS is not limited to the targeted brain region as changes can also occur at distant
interconnected sites of the brain. The application of rTMS over brain regions such as the DLPFC
implicated in the pathophysiology of neuropsychiatric disorders including schizophrenia
therefore represents a possible mechanism in which to influence subcortical regions that regulate
50
emotion and behaviour (Conca, Koppi et al. 1996; Ben-Shachar, Belmaker et al. 1997; Post and
Keck 2001; Burt, Lisanby et al. 2002; Gershon, Dannon et al. 2003).
1.7.1 Induced Cortical Changes with rTMS
Cortical changes induced by rTMS have been shown in both animal models and in humans. In
the rat brain, rTMS has been shown to induce significant changes in neuronal circuits evidenced
by changes in behaviour and an attenuation of the hypothalamic-pituitary-adrenocortical system
(Fleischmann, Prolov et al. 1995; Keck, Welt et al. 2002). Furthermore, it has been demonstrated
that rTMS increases dopamine in the dorsal hippocampus (Keck, Welt et al. 2002) and the
nucleus accumbens (Keck, Welt et al. 2002; Erhardt, Sillaber et al. 2004) using microdialysis in
rodents. Repetitive TMS in the rat has also been shown to change in the expression of proteins
reflected by the synthesis of GABA by two isoforms of GAD (Trippe, Mix et al. 2009). That is,
Trippe et al (2009) demonstrated that 1 Hz rTMS in the rat reduced expression of GAD67 and
increased the expression of GAD65 and GABA transporter (GAT-1) compared to sham
stimulation (Trippe, Mix et al. 2009). In humans, rTMS has been shown to induce changes in
cortical inhibition. For example, increased rTMS frequency up until 20 Hz applied to the motor
cortex has been shown to enhance indexes of GABAB receptor mediated inhibitory
neurotransmission (Daskalakis, Moller et al. 2006) but with 25 Hz stimulation a reduction in this
activity has also been shown (Khedr, Rothwell et al. 2007). Repetitive TMS has also been shown
to induce changes in indexes of GABAA receptor mediated inhibitory neurotransmission applied
at 5 Hz over the motor cortex (Takano, Drzezga et al. 2004) and following 10 Hz (Jung, Shin et
al. 2008). In addition, combined rTMS/PET studies in healthy subjects demonstrated that 10 Hz
rTMS over the DLPFC resulted in increased levels of extracellular dopamine (Strafella, Paus et
al. 2001), while 1 Hz rTMS resulted in increased regional blood flow in the stimulation site
(DLPFC) and in the ventrolateral prefrontal cortex (Eisenegger, Treyer et al. 2008). Taken
51
together, animal and human studies demonstrate that rTMS has the potential to change cortical
excitability through modulation of different neurotransmitters including dopamine and GABA.
1.7.2 Evidence for Working Memory Improvement in Healthy Subjects with rTMS
In healthy subjects, rTMS has been employed to investigate the roles of different brain regions in
working memory by inducing virtual lesions (For example, (Sandrini, Rossini et al. 2008)) and
also to examine if rTMS improves performance (Luber, Kinnunen et al. 2007; Hamidi, Tononi et
al. 2008; Hamidi, Tononi et al. 2009; Postle and Feredoes 2009; Preston, Anderson et al. 2009) .
Specifically, high frequency rTMS has been shown to reduce reaction times while performing
working memory tasks when applied to the parietal cortex (Luber, Kinnunen et al. 2007),
superior parietal lobule (Hamidi, Tononi et al. 2008) and to the DLPFC (Preston, Anderson et al.
2009). Improvements in working memory performance accuracy have also been reported. For
example, Feredoes and Postle (2009) applied 8 Hz rTMS over the left inferior frontal gyrus while
healthy subjects performed a working memory task (response deadline task) and observed an
improvement in the false alarm rate in response to interference probes (Postle and Feredoes
2009). Hamidi et al (2009) examined the effect of rTMS applied over the superior parietal lobule
and the DLPFC on delayed recognition and recall working memory tasks in healthy subjects.
This study reported that only 10 Hz rTMS over the right DLPFC resulted in an increase in
accuracy on the delayed recognition task with no differences found with superior parietal lobule
application (Hamidi, Tononi et al. 2009). Although there are a limited number of studies that
have demonstrated an improvement in performance with rTMS, it is possible that rTMS may
have a greater effect in patients with schizophrenia with severe deficits in working memory
function. In addition, the application of repeated sessions of rTMS may be more effective in
52
enhancing working memory performance in both healthy subjects and in patients with
schizophrenia.
1.7.3 Repetitive TMS as a Treatment for Schizophrenia
Recently, rTMS has been examined as treatment options for patients with schizophrenia owing to
the growing observation of non-responders to antipsychotic medication (Hajak, Marienhagen et
al. 2004; Jansma, Ramsey et al. 2004; Hoffman, Gueorguieva et al. 2005). In addition,
electroconvulsive therapy is proven for its anti-depressant effects however it tends to substantiate
cognitive impairments in patients with schizophrenia (Squire 1982; Sackeim, Portnoy et al. 1986;
Weiner, Rogers et al. 1986). In this regard, rTMS has been examined for its treatment effects on
auditory hallucinations, negative symptoms (Hoffman, Boutros et al. 2000; Hoffman, Hawkins et
al. 2003; Hoffman, Gueorguieva et al. 2005; Jandl, Bittner et al. 2005; Poulet, Brunelin et al.
2005) and more recently on cognitive deficits in this disorder. For example, seminal studies
conducted by Hoffman and colleagues (2000, 2003) applied 1 Hz rTMS over the left
temporoparietal cortex of 24 patients over the course of nine days. In 9 of 12 patients who
received active stimulation experienced a greater than 50% reduction in the frequency and
severity of their auditory hallucinations as compared to sham stimulation. It was also
demonstrated that their auditory hallucinations became less salient in these patients. Importantly,
52% of patients maintained improvement for up to 15 weeks (Hoffman, Hawkins et al. 2003).
Fitzgerald et al (2005) also examined the effect of 10 sessions of 1 Hz rTMS over the electrode
TP3 of the international 10-20 system for EEG placement on auditory hallucinations in patients
with schizophrenia. In this study, however, no improvements were observed in auditory
hallucination rating and no improvement or deterioration was reported in cognitive testing
(Fitzgerald, Benitez et al. 2005). There are only a couple of studies that have used rTMS to
specifically examine its effect on cognition in patients with schizophrenia. First, in a pilot study
53
based on 4 patients with schizophrenia, Sachdev et al (2005) applied 15 Hz rTMS over the left
DLPFC for 4 weeks and reported no significant improvement in general cognitive state,
executive function (working memory, verbal fluency, cognitive flexibility) or psychomotor speed
(Sachdev, Loo et al. 2005). Second, Huber et al (2003) applied 20 Hz rTMS over the left DLPFC
for 2 weeks in 12 patients with schizophrenia that resulted in an improvement in psychomotor
speed in women but not men (Huber, Schneider et al. 2003). Together these studies show some
potential with high frequency rTMS applied to the DLPFC in improving auditory hallucinations
and cognitive deficits in patients with schizophrenia; however, there is a need for more studies
that examine different rTMS parameters, session duration, the inclusion of a placebo-controlled
sham group and healthy subject group with increased sample size.
1.7.3.1 Targeting the DLPFC
As previously described the DLPFC has been implicated in the pathophysiology of schizophrenia
in addition to its critical role in working memory. As such, the DLPFC is a key brain area to be
targeted in the treatment of cognitive deficits with rTMS. Traditionally, the DLPFC has been
localized with the 5-cm rule which involves first stimulating the hot spot for a hand muscle (i.e.,
abductor pollicis brevis) through single pulse TMS and then measuring 5 cm anterior from this
position along a parasaggital line (George, Wassermann et al. 1995; Pascual-Leone, Rubio et al.
1996). Another method for determining this position is using the 10-20 EEG system (Jasper
1958) which places the coil over electrodes F3 and F4 for targeting the left and right DLPFC,
respectively (Gerloff, Corwell et al. 1997; Rossi, Cappa et al. 2001). These methods however
have been scrutinized for their poor targeting accuracies particularly in regards to inter-subject
and inter-rater variability. More recently, researchers and clinicians have turned to MRI
neuronavigational techniques to localize the DLPFC thereby decreasing inter-subject variability
with the other two methods. However, the DLPFC is a functional brain region and therefore
54
cannot be identified through anatomical inspection of a T1-weighted MRI image. In this regard,
our lab has developed a novel neuronavigational method which first estimates the DLPFC on the
cortex based on fMRI coordinates during working memory and then estimates this position on
the scalp. This method is advantageous over previous methods because it minimizes inter-subject
and inter-rater sources of variability. Furthermore, our method calculates the angle of the TMS
coil into the scalp position of the DLPFC. Finally, the advantage of our method is that it can be
used to calculate the scalp position for any functional brain region to be targeted with rTMS. The
findings of this study represent an important advancement in improving the localization of the
DLPFC to optimize rTMS as a potential treatment for cognitive deficits in schizophrenia. This
study has been published (Rusjan et al 2010) and has been included as an appendix to this thesis
(Appendix A).
1.7.4 Repetitive TMS and Cortical Oscillations
The effect of rTMS on cortical oscillations has yielded some interesting findings. Moreover,
these studies demonstrated that high frequency over the left DLPFC modulates cortical
oscillations across all frequency ranges and also produce changes in other distal brain regions.
For example, 10 Hz rTMS applied to the left DLPFC in healthy subjects was shown to increase
delta power in the frontal, central, and parietal brain regions during an EEG session with eyes
closed tested 10 minutes post rTMS (Griskova, Ruksenas et al. 2007). In this study, however,
subjects were not blind to their rTMS group assignment and although a sham placebo condition
was employed, stimulation was applied in a different brain region and at a lower intensity. In this
regard, Okamura et al (2001) conducted a study in 32 subjects that were randomized to receive
either active or sham 10 Hz rTMS applied over the left prefrontal cortex. Importantly subjects
were blind to their rTMS condition and sham stimulation was applied at the same parameters as
the active stimulation with the coil angled perpendicular to the left prefrontal cortex (Okamura,
55
Jing et al. 2001). In this study, 10 Hz rTMS applied to the left prefrontal cortex in healthy
subjects resulted in a significant increase in peak frequency of EEG across the scalp with no
change in absolute power 2 minutes following stimulation. This effect was most robust in the
frontal brain region, however, increases were also observed in the temporal region distal from the
stimulation site. By contrast, 5 minutes post rTMS stimulation significantly increased absolute
power across delta, theta, alpha, beta, and gamma frequency bands with no change in peak
frequency (Okamura, Jing et al. 2001). In a similar study using the same rTMS parameters, Jing
and Takigawa (2000) observed directed coherence between the frontal and parietal brain regions
in both hemispheres immediately following stimulation. Moreover, this effect was found to be
most significant in the alpha band with the directed coherence from the frontal to the parietal
brain region increasing by 32% even cross hemispheres. This finding demonstrates that distal
brain regions can be modulated by rTMS even if they are not directly connected (Jing and
Takigawa 2000). It is possible that nonspecific thalamic system and the corpus callosum may be
important in mediating the transhemispheric signal transmission (Hamada and Wada 1998),
while the brainstem may play a role in the spread of the signals (McIntyre and Goddard 1973;
Kievit and Kuypers 1975). It has also been suggested that the effect of rTMS on oscillatory
activity may result from a direct cortical effect (Pascual-Leone, Houser et al. 1993). Specifically,
Pascual-Leone and colleagues propose that the excitatory intracortical axons are collateral to
pyramidal cells and inhibitory interneurons. Furthermore, inhibitory interneurons form feedback
loops and project to the pyramidal cells. The balance between excitatory and inhibitory activity
is critical in maintaining homeostasis under normal circumstances, however, rTMS upsets this
balance with the accumulation of EPSPs in the absence of IPSPs due to the difference in the
number of synapses and conduction along myelinated monosynaptic excitatory collaterals
(Pascual-Leone, Houser et al. 1993). This proposed mechanism however does not account for
56
changes found in the Jing and Takigawa study that demonstrated immediate changes in peak
frequency that suggests another mechanism may be involved (Jing and Takigawa 2000).
1.8 Outline of Experiments
Chapter 2 will provide a background and rationale for the three studies conducted as part of the
PhD program. In this chapter the objectives and hypothesis for these experiments will be
outlined. Chapters 3, 4 and 5 represent studies one, two, and three, respectively. These studies
represent publications from the work done during the PhD program. Studies one and two have
been published. Study three has been submitted for publication. Because these studies have been
submitted as standalone articles, material contained in each of these articles may overlap with
material presented in other articles in addition to material that was discussed in the Introduction.
57
2 Overview of Experiments, Hypotheses and Objectives
2.1 Study 1: Evidence for Excessive Frontal Evoked Gamma Oscillatory Activity in Schizophrenia during Working Memory
2.1.1 Background
Oscillatory activity is associated with working memory. Specifically, increases in gamma (30-50
Hz) oscillatory activity has been demonstrated in response to increases in working memory load
in epileptic patients (Howard, Rizzuto et al. 2003) and in healthy subjects (Basar-Eroglu, Brand
et al. 2007). Furthermore, Basar-Eroglu et al (2007) reported that patients with schizophrenia
generate increased gamma oscillatory activity when tested in the N-back at 3 levels (0-2-back) of
working memory load. In addition, a relationship between oscillatory activity generated during
working memory and clinical symptoms has yet to be demonstrated. The aim of the first study of
the thesis, therefore, was twofold: 1) to quantify alterations in frontal evoked oscillatory activity
across the 5 frequency bands (i.e., delta, theta, alpha, beta and gamma) in patients with
schizophrenia compared to healthy subjects while performing the N-back task administered at 3
levels (1-3-back); 2) to examine the relationship between gamma oscillatory activity, working
memory performance, and the severity of clinical symptoms in patients with schizophrenia. The
study included 24 medicated patients with schizophrenia or schizoaffective disorder confirmed
by the DSM-IV, and 24 healthy subjects. All subjects were right handed (Oldfield 1971).
Patients were recruited through the Schizophrenia and Continuing Care Program at the Centre
for Addiction and Mental Health (CAMH), while healthy subjects were recruited through poster
advertisement placed at CAMH and the University of Toronto.
58
2.1.2 Primary Objectives
First, we will compare frontal evoked oscillatory activity in patients with schizophrenia
compared to healthy subjects while performing the N-back task. Second, we will examine the
relationship between gamma oscillatory activity, performance, and severity of clinical symptoms
in patients with schizophrenia.
2.1.3 Primary Hypotheses
We first hypothesized that patients with schizophrenia will exhibit altered frontal evoked gamma
oscillatory activity while performing the N-back task across working memory load. Second, we
hypothesized that altered gamma oscillatory activity will be related to impaired working memory
performance and severity of clinical symptoms.
2.1.4 Results
We found that patients with schizophrenia performed significantly worse in the N-back task
compared to healthy subjects that was accompanied by alterations in oscillatory activity.
Specifically, we confirmed the findings of Basar-Eroglu (2007) and demonstrated that patients
with schizophrenia, compared to healthy subjects, generate excessive frontal evoked gamma
oscillatory (F=4.561, p=0.039) regardless of working memory load. Furthermore, we found that
this effect was still present at equivalent performance levels (t=-2.898, p=0.006) (Barr, Farzan et
al. 2010). Finally, we demonstrate an inverse relationship between working memory
performance and negative but not positive symptoms (r=-0.416, p=0.043). These results suggest
that gamma oscillatory activity is excessive in patients with schizophrenia and is not modulated
by working memory load in contrast to healthy subjects.
The study raised two important questions. First, can high frequency rTMS modulate frontal
evoked gamma oscillations generated during the N-back in healthy subjects and in patients with
59
schizophrenia? Second, if rTMS indeed modulate frontal evoked gamma oscillatory activity,
would these neurophysiological changes be reflected in performance improvement? Study two of
this thesis was intended to measure the effect of high frequency rTMS applied to the DLPFC on
frontal evoked gamma oscillatory activity generated during the N-back in healthy subjects. Study
three of this thesis then looked to examine if excessive frontal evoked gamma oscillations in
patients with schizophrenia observed in study one could be inhibited by rTMS applied to the
DLPFC.
2.2 Study 2: Potentiation of Gamma Oscillatory Activity through Repetitive Transcranial Magnetic Stimulation (rTMS) of the Dorsolateral Prefrontal Cortex
2.2.1 Background
From study one it was observed that patients with schizophrenia generate excessive frontal
evoked gamma oscillations while performing the N-back task. Furthermore, this effect was
accompanied by deficits in working performance which was then found to be related to the
severity of negative symptoms. It is therefore important to explore whether frontal gamma
oscillations can be influenced with high frequency rTMS applied over the DLPFC.
Repetitive (TMS) has been previously shown to improve cognition in other patient populations.
For example, improvements in language function with rTMS has been demonstrated in healthy
subjects (Sparing, Mottaghy et al. 2001) and in patients with major depressive disorder (Little,
Kimbrell et al. 2000; Martis, Alam et al. 2003; O'Connor, Jerskey et al. 2005). In addition, high
frequency rTMS has been shown to enhance GABAergic inhibitory neurotransmission in healthy
subjects (Daskalakis, Moller et al. 2006). Given the evidence for the role of GABAergic
inhibitory neurotransmission in generation (Whittington, Traub et al. 1995; Wang and Buzsaki
1996; Bartos, Vida et al. 2007) and modulation (Whittington, Traub et al. 1995; Brown, Davies
60
et al. 2007; Leung and Shen 2007) of gamma oscillations, rTMS may exert its influence on
gamma oscillatory activity through the modulation of GABAergic activity.
This study included 22 right-handed (confirmed with Oldfield Handedness Inventory (Oldfield
1971) healthy subjects comprised of equal males and females. Subjects were recruited through
poster advertisement placed at CAMH and the University of Toronto. This was a blind, placebo-
controlled pre-post study design. Subjects were randomly assigned to receive either active or
sham (placebo) rTMS stimulation and were blind to their group assignment. All subjects
completed the N-back task one week before rTMS was applied to the DLPFC followed by a
second testing of the N-back task. The effect of rTMS on frontal evoked oscillatory activity
during the N-back task was measured with EEG.
2.2.2 Primary Objective
We will evaluate the effect of 20 Hz rTMS applied bilaterally to the DLPFC on frontal evoked
gamma oscillatory activity measured during the N-back in healthy subjects.
2.2.3 Primary Hypothesis
We hypothesize that active rTMS would enhance frontal evoked gamma oscillatory activity in
response to increased working memory load and this effect would be absent in activities in the
other frequency bands (i.e., delta, theta, alpha, and beta).
2.2.4 Results
Our results demonstrate that rTMS significantly increases gamma oscillations generated during
the N-back task in healthy subjects (Barr, Farzan et al. 2009). That is, active rTMS results in
greater gamma oscillatory activity compared to baseline (prior to rTMS) (p<0.0001) and to sham
stimulation (p=0.0028). Moreover, these effects were most pronounced at higher working
61
memory loads (2-back: p=0.0032; 3-back: p=0.0288). Importantly, the effect of active rTMS on
gamma oscillatory was found to be selective to activity in the gamma frequency range and were
isolated to the frontal brain region when compared to the posterior brain region (t=4.101,
p<0.005). These results therefore demonstrate that rTMS applied to the DLPFC can modulate
gamma oscillatory activity a finding that could be translated into patients with schizophrenia.
2.3 Study 3: The Effect of Transcranial Magnetic Stimulation on Gamma Oscillatory Activity in Schizophrenia
2.3.1 Background
Study one of this thesis demonstrated that patients with schizophrenia generate excessive gamma
oscillatory activity while performing the N-back task compared to healthy subjects. Study two
demonstrated that high frequency rTMS applied bilaterally to the DLPFC potentiates frontal
evoked gamma oscillatory activity in healthy subjects that most pronounced in the N-back
conditions with the greatest cognitive demands. The findings of these two studies suggest that
rTMS over the DLPFC may potentially modulate excessive gamma oscillatory activity observed
in patients with schizophrenia while performing the N-back task.
This study tested 24 medicated patients with a diagnosis of schizophrenia or schizoaffective
disorder confirmed by the Structural Clinical Interview for DSM-IV (Spitzer 1994). Patients
were recruited through the Schizophrenia and Continuing Care Program at CAMH. The healthy
control group consisted of 22 subjects who were recruited through poster advertisement placed at
CAMH and the University of Toronto. All subjects were right-handed (Oldfield 1971). This
study was a randomized, double-blind placebo-controlled design. All subjects were allocated to
62
receive active or sham stimulation and the subjects and raters were blind to their rTMS
condition.
2.3.2 Primary Objective
The aim of this study was to examine the effect of high frequency rTMS applied bilaterally to the
DLPFC on frontal gamma oscillatory activity generated during the N-back in patients with
schizophrenia compared to healthy subjects.
2.3.3 Primary Hypothesis
It was hypothesized that rTMS would inhibit excessive gamma oscillatory activity in patients
with schizophrenia compared to sham stimulation and in contrast to healthy subjects owing to the
fact that rTMS has been previously shown to result in the potentiation of GABA inhibitory
neurotransmission in healthy subjects (Daskalakis, Moller et al. 2006; Jung, Shin et al. 2008).
63
3 Study One: Evidence for Excessive Frontal Evoked Gamma Oscillatory Activity in Schizophrenia during Working Memory
3.1 Abstract
Gamma (γ) oscillations (30-50 Hz) elicited during working memory (WM) are altered in
schizophrenia (SCZ). However, the nature of the relationship between evoked frontal oscillatory
activity, WM performance and symptom severity has yet to be ascertained. This study had two
objectives. First, to extend previous studies by examining delta, theta, alpha, beta, and gamma
(δ, , , , and γ) oscillatory activities during the N-back task in SCZ patients compared to
healthy subjects; second, to evaluate the relationship between oscillatory activities elicited during
the N-back, performance, and clinical symptoms in SCZ patients. Patients with SCZ elicited
excessive frontal γ oscillatory activity that was most pronounced in the 3-back condition
compared to healthy subjects. Reduced frontal activity at all WM loads was also observed in
patients with SCZ compared to healthy subjects. Task performance was inversely correlated with
negative symptoms but not with positive symptoms. Our findings suggest that evoked frontal
oscillatory activities during WM are selectively altered in the γ and frequency bands that may
contribute to WM impairment in SCZ patients. These findings may provide important insights
into the pathophysiology underlying WM deficits, its relationship to negative symptoms and may
represent a potential neurobiological marker for cognitive enhancing strategies in SCZ.
3.2 Introduction
Gamma (γ) oscillations (30-50 Hz) have been shown to be a key neurophysiological mechanism
underlying working memory (WM), a cognitive process involving the maintenance and
64
manipulation of information on-line (Baddeley 1986). Increased γ oscillatory activity with
increased WM load has been demonstrated in epileptic patients (Howard, Rizzuto et al. 2003)
and healthy subjects (Basar-Eroglu, Brand et al. 2007) with electroencephalography (EEG)
especially in the dorsolateral prefrontal cortex (DLPFC). Although γ modulation in the DLPFC
is most consistently associated with WM, delta (δ; 1-3.5 Hz), theta (; 4-8 Hz), alpha (; 9-12
Hz), and beta (; 14-28 Hz) activities also vary with WM load (Barr, Farzan et al. 2009).
Altered oscillatory activity during WM has been demonstrated in SCZ. Abnormal δ (Ince,
Pellizzer et al. 2009), (Schmiedt, Brand et al. 2005; Haenschel, Bittner et al. 2009),
(Bachman, Kim et al. 2008; Stephane, Ince et al. 2008; Ince, Pellizzer et al. 2009) and
(Stephane, Ince et al. 2008) activation during WM has been shown in SCZ patients. However,
inconsistent reports reveal both similar (Haenschel, Bittner et al. 2009) and abnormally increased
(Basar-Eroglu, Brand et al. 2007) γ oscillatory activity during WM in SCZ compared to healthy
subjects. Furthermore, Basar-Eroglu et al (2007) demonstrated that SCZ patients failed to
modulate γ oscillatory activity with WM load as found in healthy subjects (Basar-Eroglu, Brand
et al. 2007). The lack of γ modulation with WM load in SCZ patients may be attributed to
alterations in γ –aminobutyric acid (GABA) interneurons in the DLPFC (Akbarian, Kim et al.
1995; Benes and Berretta 2001; Lewis, Hashimoto et al. 2005; Hashimoto, Bazmi et al. 2008)
critical to γ oscillatory activity. Considering the importance of GABA in the generation
(Whittington, Traub et al. 1995; Wang and Buzsaki 1996; Bartos, Vida et al. 2007) and
synchronization (Whittington, Traub et al. 1995; Brown, Davies et al. 2007) of γ oscillations,
GABAergic impairments in patients with SCZ may underlie altered modulation of γ oscillatory
during WM.
65
The relationship between γ oscillatory activity, WM performance, and clinical symptoms has yet
to be determined. In a behavioural study, Park et al (1999) reported on a relationship between
WM performance and negative symptoms in SCZ patients, such that increased negative
symptoms related to poor WM (Park, Puschel et al. 1999). Additionally, a fMRI study in SCZ
patients demonstrated a correlation between strength of prefrontal-parietal connectivity and WM
performance, while positive symptoms related to decreased connectivity (Henseler, Falkai et al.
2009). Relationships have also been uncovered between γ oscillatory activity with both positive
and negative symptoms although through an auditory odd-ball task and not WM (Lee, Williams
et al. 2003). In this study, patients were divided by symptom predominance and found that
positive symptoms correlated with increased γ, while negative symptoms were related to a
reduction of γ (Lee, Williams et al. 2003). As such, γ oscillations during WM may be related to
both performance and symptom severity in SCZ patients. In the current study, therefore, we
aimed to extend previous findings by examining evoked oscillatory activity during WM across δ,
, , , and γ frequency components measured from the frontal region in SCZ patients compared
to healthy subjects. We also sought to examine if neuronal oscillations during WM were related
to task performance and symptom severity.
3.3 Materials and Methods
3.3.1 Subjects
Twenty-four (males=14; females=10) patients with a diagnosis of SCZ or schizoaffective
disorder (SCZ=19; schizoaffective disorder=5), confirmed by the Structured Clinical Interview
for DSM-IV (Spitzer 1994), and 24 (males=13; females=11) healthy subjects participated in this
study. All subjects were right handed (Oldfield 1971). SCZ patients were treated with
antipsychotic medication (15.0 ± 14.1 mg olanzapine, 5 patients; 318.8 ± 196.3 mg clozapine, 8
66
patients; 3.7 ± 1.5 mg risperidone, 3 patients; 650.0 ± 378.6 mg of quetiapine, 4 patients; 0.7 ±
0.8 mg of fluphenazine, 2 patients ; 5 mg haloperidol, 1 patient; 15 mg aripiprazole, 1 patient;
dose/day). Severity of psychopathology was evaluated using the positive and negative symptom
scale (PANSS; (Kay, Fiszbein et al. 1987)), scale for the assessment of negative symptoms
(SANS; ((Andreasen 1989)) and the Calgary Depression Scale (CDS; ((Addington, Addington et
al. 1993); Table 1). Subject groups were similar in age (t(45)=0.320, p=0.750), but differed in
education (independent t-tests: t(45)=14.771, p<0.001; Table 1). Exclusion criteria for all
subjects included a history of substance abuse or dependence in the last 6 months determined
through the DSM-IV or pregnancy. Healthy subjects were excluded if a concomitant major and
unstable medical, neurologic illness and/or the presence of psychopathology determined by the
personality assessment screener (PAS; Psychological Assessment Resources, Inc). Subjects
provided their written informed consent and the protocol was approved by the Centre for
Addiction and Mental Health in accordance with the declaration of Helsinki.
3.3.2 N-Back Task
Subjects performed the N-back task while EEG activity was recorded (STIM2, Neuroscan,
U.S.A.). Stimuli were presented on a computer monitor one at a time and participants were
required to push one button (target) if the present stimulus was identical to the stimulus
presented ―N‖ trials back; otherwise, subjects pushed a different button (non-target). Thus, the
effect of increasing cognitive demand on γ oscillatory activity was tested by varying the ―N‖ in
the 1-, 2- and 3-back conditions. Stimuli consisted of black capital letters presented for 250
msec followed by a delay period of 3000 msec during which the subject was required to respond
(Figure 1). In the 1-, and 2-back, stimuli were presented continuously for 15 minutes and for 30
minutes in the 3-back. The 3-back was administered for double the length of time to ensure a
satisfactory number of correct responses were contained for the data analysis (Table 2). The
67
number of target letters in each condition was: 46 of 198 (23.2%) 1-back; 31 of 197 trials
(15.7%) 2-back, 59 of 400 trials (14.6%) 3-back condition. The N-back took 1 hour to complete
with the order of conditions randomized within subjects and counterbalanced across subjects to
prevent order effects.
3.3.3 EEG Measurement of Evoked γ Oscillatory Activity
Evoked oscillatory responses are phase-locked to the stimulus onset with a fixed latency within
the first 100 msec following stimulus onset and are measured by stimulus-triggered averaging of
responses (Tallon-Baudry and Bertrand 1999). While, induced oscillatory activity is not phase-
locked to stimulus onset and jitters in latency varying from trial to trial; thus, responses are
cancelled out when averaged (Tallon-Baudry and Bertrand 1999). Although both evoked and
induced γ activities have been associated with WM (Howard, Rizzuto et al. 2003; Basar-Eroglu,
Brand et al. 2007; Barr 2008), induced γ activity has been suggested to reflect microsaccadic eye
movement rather than neuronal processing during cognitive paradigms (Yuval-Greenberg,
Tomer et al. 2008). Additionally, γ activity during cognitive tasks has been shown to reflect
cranial musculature artifact (Shackman 2010), however, since this activity is characterized by
irregular spikes and waves present in all spectral frequencies such artifact should be significantly
reduced when multiple trials are averaged with evoked analytical methods (Barr and Daskalakis
2010). As such, we measured mean evoked γ power from frontal electrodes while subjects
completed the N-back task.
3.3.4 EEG Recording
EEG data were acquired using a 64-electrode cap and Synamps2 DC-coupled EEG system
(Compumedics, U.S.A.). Four electrodes placed on the outer side of each eye, above, and below
the left eye to monitor eye movement artifact. Data was recorded at a rate of 1000 Hz DC and
68
with a 0.3 to 200 Hz band pass hardware filter. Electrode impedances were lowered to < 5 kΩ.
All channels were referenced to an electrode placed posterior to the Cz electrode.
3.3.5 Offline EEG processing
Data was filtered off-line using a 1 to 100 Hz band pass zero phase shift filter (slope, 24 dB/oct).
Epochs were defined as –1000 to +3095 msec relative to the cue onset and were baseline
corrected with respect to the prestimulus interval (-1000 to cue onset). Trials were manually
inspected and any error trials or epochs containing artifact (movement or electrooculogram
exceeding +/- 50 µV) were excluded from further analysis. On average, we excluded 26% of all
correct trials (TC+NTC) from the healthy subject group and 27% from the SCZ patient group.
Finally, oscillatory power was extracted by decomposing the signals by means of a hamming-
based zero-phase shift finite impulse response filter.
3.3.6 Data Analysis
The total number of correct trials (target correct (TC) and non-target correct (NTC)) including
those trials rejected due to artifact were included in the data analysis for WM performance and
reaction time. Two separate repeated measures ANOVA for WM and reaction were performed
with Group (SCZ versus healthy subjects) as a between-subject factor with WM load (1 versus 2
versus 3) as a within-subject factor, significance level set at p<0.05. The assumption of
sphericity was met with each repeated measures ANOVA. Subsequent pairwise comparisons
were performed with the level of significance Bonferroni-adjusted (SPSS 15.0, SPSS Inc.
Chicago, Illinois, USA).
Artifact-free EEG data were imported into MATLAB (The MathWorks, Inc. Natick, MA, USA)
using the EEGLAB toolbox (Delorme and Makeig, 2004) for subsequent analysis. Evoked γ
power (30-50 Hz) was averaged over the delay period for TC and NTC responses for each WM
69
load. Mean evoked power during these responses (TC + NTC) in the frontal electrodes (AF3/4,
F5/6, F3/4, F1/2, and FZ) were measured and then averaged. Since spectral analysis of EEG
activity is often not normally distributed (Bender, Schultz et al. 1992), the data was log
transformed prior to analysis. For each frequency (δ, 1-3.5 Hz; , 4-8 Hz; , 9-12 Hz; , 14-28
Hz; and γ, 30-50 Hz), a repeated measures analysis of variance (ANOVA) was performed with
Group (SCZ versus healthy subjects) as a between-subject factor with WM load (1 versus 2
versus 3) as a within-subject factor, significance level set at p<0.05. The assumption of
sphericity was met with each repeated measures ANOVA. Subsequent pairwise comparisons
were performed with the level of significance Bonferroni-adjusted. Spearman ranked correlations
were used to evaluate the relationship between γ oscillatory activity, performance and clinical
symptoms in SCZ patients (SPSS 15.0, SPSS Inc. Chicago, Illinois, USA).
3.4 Results
3.4.1 Working Memory Performance
N-back performance decreased with increased WM load (F(2,104)=69.678; p<0.0001; Table 3).
Subjects performed significantly worse in 3- compared to 2- and in 2- compared to 1-back
condition (p<0.0001 in both cases). The Group difference in N-back performance was also
significant (F(1,44)=9.533; p=0.004) with SCZ patients performing significantly worse compared
to healthy subjects. Mean reaction time increased with increased WM load (F(2,88)=17.263;
p<0.0001; Table 3) that significantly increased from the 1- to 2-back (p< 0.0001) and trended
towards significance from the 2- to the 3-back condition (p=0.092).
3.4.2 Evoked γ Power
The effect of WM load on mean γ power was significant (F(2,84)=3.820; p=0.026) and was
greatest in the 2-back (Figure 2A). The Group x WM load interaction was also significant (F(2,84)
70
= 10.537; p<0.001), independent t-tests revealed that SCZ patients elicited higher γ power at
each WM load that reached significance in the 3-back condition (p<0.001; shown as
Topographical plots Figure 2B). In healthy subjects, γ power increased from 1- to the 2-back
and decreased in the 3-back (p=0.001 in each case). Conversely, no differences were observed
between N-back conditions in SCZ patients (p>0.05 in each case), reflecting a lack of γ
modulation with WM load. The Group effect was significant (F(1,42) = 4.561; p=0.039) with SCZ
patients generating excessive γ activity overall. At equivalent performance levels, γ oscillatory
activity was compared between the two groups. An independent t-test first determined that SCZ
1-back performance was equivalent to 3-back performance healthy subjects (t(44)=-0.941,
p=0.352). Next, γ activities were compared in these conditions and found that SCZ patients still
elicited significantly greater γ oscillatory activity compared to healthy subjects at equivalent
performance levels (t(45)=-2.898, p=0.006). SCZ patients, therefore, elicited excessive γ
oscillatory activity independent of WM load compared to healthy subjects.
3.4.3 Spectral Analysis of Other Frequency Bands
To test whether alterations in SCZ patients was selective to the γ, we examined oscillatory power
in the following four bands: δ, , , and compared to the healthy subjects. No group
differences were found in δ, , and ; however, SCZ patients elicited significantly reduced
activity compared to healthy subjects (F(1,40)=15.859, p0.0001) that did not vary with WM load
(Figure 3).
3.4.4 Relationship Between Evoked Power, N-back Performance, and Symptom Severity
Spearman ranked correlations were performed to examine the relationship between γ power in
the 3-back with performance and clinical symptoms owing to the fact that group differences in γ
71
power in this condition were most pronounced. An inverse relationship was found between
negative symptoms measured by the SANS and working memory performance (r=-0.538,
p=0.007). That is, the greater the severity of negative symptoms the worse SCZ patients
performed in the 3-back condition.
3.4.5 Effect of Antipsychotic Medication
A Pearson correlation coefficient was performed to determine if excessive γ and reduced
oscillatory activity in the 3-back was related to antipsychotic medication using chloropromazine
equivalents (CPZ; (Woods 2003)). No relationships were found between γ or oscillatory
activities and CPZ equivalents in the 3-back. Further, no relationship was observed between 3-
back performance and CPZ equivalents examined through Pearson correlation coefficient.
3.5 Discussion
SCZ patients generated excessive evoked frontal γ oscillatory activity that was most pronounced
in the 3-back compared to healthy subjects. Task performance was inversely correlated with
negative symptoms but not with positive symptoms. Finally, there was a reduction in
oscillatory activity in SCZ patients, but this was not related to symptom severity, task
performance or WM load.
Consistent with Basar-Eroglu et al (2007), we observed an increase in frontal γ oscillatory
activity at each WM load, particularly in the 3-back condition in SCZ patients compared to
healthy subjects. Although in the Basar-Eroglu study, the N-back task was administered up until
the 2-back condition, we observed the most pronounced effects in the 3-back condition. Such an
effect is, in part, driven by the decrease in frontal γ oscillatory activity in the 3-back relative to
the 2-back condition in healthy subjects. This inverted u-shaped curve in response to increased
72
WM load is consistent with fMRI studies (Callicott, Mattay et al. 1999) reflective of poorer WM
performance possibly due to diminished attentional resources (Cowan 2001; Kane, Bleckley et
al. 2001; Wheeler and Treisman 2002). Nevertheless, we observed excessive frontal γ
oscillatory activity in SCZ patients regardless of WM load suggesting a lack of γ modulation in
response to increased cognitive demand. This finding is further substantiated by the fact that
SCZ patients‘ frontal γ oscillatory activity is significantly greater compared with healthy subjects
at equivalent performance levels and, therefore, was not due to poorer performance. Possibly
this phenomenon is related to altered GABAergic inhibitory neurotransmission in the DLPFC.
There are, in fact, several lines of evidence to support this contention. First, post-mortem studies
have demonstrated reductions in the 67 kilodalton isoform of glutamic acid decarboxylase
(GAD67) (Akbarian, Kim et al. 1995; Benes and Berretta 2001; Volk, Pierri et al. 2002; Lewis,
Hashimoto et al. 2005) and calcium binding protein parvalbumin in the DLPFC (Collin, Chat et
al. 2005). Second, reduction of parvalbumin leading to increased facilitation of GABA release
during repetitive synaptic firing has been directly linked modulation (Vreugdenhil, Jefferys et
al. 2003; Sohal, Zhang et al. 2009). Third, a recent study measured neurophysiological indices
of GABAB receptor inhibitory neurotransmission from the DLPFC in SCZ compared to patients
with bipolar disorder and healthy subjects through combined TMS-EEG (Farzan, Barr et al.
2010). Waveforms were decomposed into frequency components and demonstrated that
inhibition of oscillations were significantly reduced in DLPFC in SCZ compared to the other
two groups (Farzan, Barr et al. 2010). It is possible that the lack of inhibition of oscillatory
activity in SCZ patients results in excessive oscillations during WM that may contribute to
performance deficits in this disorder.
73
Consistent with previous studies, we observed reduced β oscillatory activity in SCZ patients.
For example, SCZ studies have demonstrated β band reductions in phase synchrony in visual
perception of Mooney faces (Uhlhaas, Linden et al. 2006), inter-trial coherence preceding speech
generation (Krishnan, Vohs et al. 2005), and in steady-state visual evoked potentials (Krishnan,
Vohs et al. 2005). Such altered β oscillations may be related to GABAB activity, as GABAB
agonist L-baclofen administered at low doses (5mg/kg) increases β and decreases γ power, while
at higher doses (50 mg/kg) a reduction in β and an increase in γ power was observed in mice
(Marrosu, Santoni et al. 2006). The differential effect of L-baclofen on γ and β oscillatory
activity their study is congruent with the finding that such activities are synchronized by different
cellular mechanisms (Traub, Whittington et al. 1999). Furthermore, studies have reported on the
phenomenon of a γ to β frequency shift in animal models (Whittington, Stanford et al. 1997) and
in SCZ (Hong, Summerfelt et al. 2004). Specifically, Hong et al (2004) examined the
suppression of auditory evoked potentials using the P50 paradigm in SCZ patients compared to
healthy subjects. This paradigm involves two auditory clicks (S1, S2), and with normal sensory
gating the magnitude of the P50 response to S2 is attenuated compared to S1 response. In
healthy subjects, a shift from γ to β activity in response to S1 was found to underlie the
suppression of S2 (Hong, Summerfelt et al. 2004). That is, the level of β activity following S1,
the greater the P50 response. By contrast, the shift from γ to β activity following S1 was reduced
in SCZ patients leading to decreased P50 responses (Hong, Summerfelt et al. 2004). Together
these studies support the coupling of γ and β activities that may explain the associated elevation
in γ and reduction in β power elicited during WM in this study.
In SCZ patients, N-back performance was inversely related to the severity of negative symptoms.
This finding is consistent with several cross-sectional studies that have demonstrated a
correlation between cognitive performance and negative symptoms (Harvey, Howanitz et al.
74
1998; Park, Puschel et al. 1999). Moreover, it has been argued that SCZ patients with the
greatest cognitive impairment present the most prominent negative symptoms (Basso et al. 1998;
Villalta-Gil et al. 2006). It has further been suggested that negative symptoms mediate the
relationship between cognition and functional outcome (Ventura et al 2009).
To our knowledge, this is the first study that has shown excessive γ with a concomitant reduction
in β oscillatory activity during WM in SCZ. Furthermore, excessive γ oscillations were related
to depressive symptoms, while N-back performance was related to negative symptoms.
However, this study is limited in several ways. First, our finding of excessive γ oscillations in
SCZ may be attributed to the effects of antipsychotic medication. In this regard, Hong et al
(2004) reported enhanced 40 Hz oscillations in SCZ patients on second generation compared
those taking conventional antipsychotics (Hong et al. 2004). In their study, oscillations were
parsed into 20, 30 and 40 Hz; however, 30-50 Hz range is most conventionally examined during
cognitive tasks (Howard, Rizzuto et al. 2003). Similarly, a differential effect of antipsychotic
medication on cognitive performance has also been reported with SCZ patients on second
generation antipsychotics performing better than those patients on conventional antipsychotics
on a variety of cognitive tests, including WM (Meltzer and McGurk 1999; Bilder, Goldman et al.
2002; Sharma, Hughes et al. 2003). In our sample, 5 subjects were on conventional
antipsychotics, but no differences were found in γ or WM performance compared to patients on
second generation antipsychotics. Furthermore, there was no relationship between γ, β or
performance in the 3-back condition with CPZ equivalents. Second, the N-back task does not
allow for examination of oscillatory activity within the different sub-processes in WM (i.e.,
encoding, maintenance, retrieval). In this regard, Haenschel et al. (2009) reported reductions in
evoked , , and activities during encoding, a peak shift of induced γ activity during
75
maintenance, and reduced induced and γ activities during retrieval (Haenschel, Bittner et al.
2009). However, no differences in evoked or induced γ oscillatory activity in the frontal region
were observed in this study.
In summary, alterations in oscillatory activity during WM are selective to the γ and activation
independent of WM load in SCZ. N-back performance was also related to greater negative
symptom severity. These findings may provide important insights into the pathophysiology
underlying WM deficits, its relationship to clinical symptoms and may represent a potential
neurobiological marker for cognitive enhancing strategies in SCZ.
76
Table 1. Demographic data (±) 1 standard deviation in healthy subjects (HS) versus patients
with schizophrenia (SCZ) and the assessment of psychotic symptoms in SCZ patients.
HS SCZ
Age 37.71 (±) 10.12 37.09 (±) 11.04
Age Range 24-60 23-57
Female (n) 11 10
Male (n) 13 14
PANSS Scores Positive NA 15.79 (±) 6.87
Negative NA 15.58 (±) 8.49
Global NA 25.83 (±) 9.43
Total NA 55.50 (±) 20.71
CDS Score Total NA 2.92 (±) 3.16
SANS Score Total NA 38.54(±) 22.40
77
Table 2. Working memory (WM) behavioural data in healthy subjects (HS) versus patients with
schizophrenia (SCZ) in the 1-, 2-, and 3-back task conditions (±) 1 standard deviation.
Behavioural data was calculated on all trials that were correct including those trials that were
rejected due to artifact.
WM Behavioural Data N-Back Condition HS SCZ
WM Score (%Correct) 1-Back 90.79 (±) 7.51 79.19 (±) 19.50
2-Back 80.80 (±) 19.99 70.29 (±) 20.53
3-Back 73.56 (±) 21.08 55.03 (±) 19.23
Reaction Time (msec) 1-Back 729.50 (±) 162.74 743.04 (±) 311.66
2-Back 904.11 (±) 255.50 859.63 (±) 333.73
3-Back 981.08 (±) 293.34 870.46 (±) 324.32
78
Figure 1. A) A representation of the 1-, 2- and 3-back conditions that were completed in a
randomized order by patients with schizophrenia (SCZ) and healthy subjects. Subjects were
required to push one button (target) if the current letter was identical to the letter presented ―N‖
trials back; otherwise the participants pushed a different button (non-target). Correct responses
for target (TC) and non-target (NTC) were included in the data analysis. B) The timing of one
trial from the presentation of a one letter separated by a (+) sign followed by a subsequent letter
for a total time of 3000 msec.
79
Figure 2. A) Mean log transformed γ (30-50 Hz) oscillatory power (TC+NTC) elicited across N-
back task conditions in patients with schizophrenia (SCZ) compared to healthy subjects (HS).
Error bars represent (±) 1 standard error (SE). B) Topographical illustration of mean absolute γ
power elicited during 3-back in patients with SCZ compared to HSs. Greater γ power is
represented by hot colours.
80
Figure 3. Mean log transformed frequency power in the δ (1-3.5 Hz), (4-7 Hz), (8-12 Hz),
and (12.5-28 Hz) ranges (TC+NTC) elicited across N-back task conditions in patients with
schizophrenia (SCZ) compared to healthy subjects (HS). Error bars represent (±) 1 standard error
(SE).
81
4 Study 2: Potentiation of Gamma Oscillatory Activity through Repetitive Transcranial Magnetic Stimulation (rTMS) of the Dorsolateral Prefrontal Cortex
4.1 Abstract
Neuronal oscillations in the gamma (γ) frequency range (30-50 Hz) have been associated with
cognition. Working memory (WM), a cognitive task involving the on-line maintenance and
manipulation of information, elicits increases in γ oscillations with greater cognitive demand,
particularly in the dorsal lateral prefrontal cortex (DLPFC). The generation and modulation of γ
oscillations have been attributed to inhibitory interneuron networks that use γ –aminobutyric acid
(GABA) as their principle neurotransmitter. Repetitive transcranial magnetic stimulation
(rTMS) represents a non-invasive method to stimulate the cortex that has been shown to modify
cognition and GABA inhibitory mechanisms, particularly with higher frequencies (i.e., 10-20
Hz). We measured the effect of high-frequency rTMS over the DLPFC on γ oscillations elicited
during the N-back WM task in healthy subjects. Active rTMS significantly increased γ
oscillations generated during the N-back conditions with the greatest cognitive demand. Further,
no significant changes were found in other frequency ranges, suggesting that rTMS selectively
modulates γ oscillations in the frontal brain regions. These findings provide important insights
into the neurophysiological mechanisms that underlie higher order cognitive processes, and
suggest that rTMS may be used as cognitive enhancing strategy in neuropsychiatric disorders
that suffer from cognitive deficits.
82
4.2 Introduction
The importance of gamma (γ) oscillatory activity in higher cognitive tasks is an area of great
interest and has recently been shown as a key neurophysiological mechanism underlying
working memory (WM). Commonly defined as the ability to maintain and manipulate
information over short periods of time (Baddeley 1986), WM is often argued as the essence of all
prefrontal functions due to its importance in everyday complex cognitive tasks, such as language,
comprehension, learning, and reasoning (Baddeley 1992; Baddeley 2000). Moreover, the
dorsolateral prefrontal cortex (DLPFC) is consistently reported to mediate WM processes
revealed through enhanced blood oxygen level dependent (BOLD) activity in functional
magnetic resonance imaging (fMRI) studies (Owen 1997; Petrides 2000).
Several cognitive paradigms are used to index WM. For example, the N-back task requires
subjects to determine if the current stimulus is the same stimulus that was presented ―N‖ trials
back, thus allowing the evaluation of increasing cognitive demand (i.e., WM load) on γ
oscillatory activity. Further, a growing body of evidence has demonstrated increased γ
oscillatory activity with cognitive demand (Howard, Rizzuto et al. 2003; Basar-Eroglu, Brand et
al. 2007; Meltzer, Zaveri et al. 2008). For example, Cho and colleagues (2006) reported that γ
oscillatory activity increased with increased cognitive control in the frontal regions that was
related to performance (Cho, Konecky et al. 2006) suggesting that γ is modulated by cognitive
demand, which may underlie WM performance. It has been proposed that -aminobutyric acid
(GABA) inhibitory interneurons in the DLPFC contribute to the generation and synchronization
of pyramidal neurons necessary for optimal WM performance (Wang and Buzsaki 1996; Traub,
Michelson-Law et al. 2004). For instance, Wilson and colleagues (1994) reported fast-spiking
GABAergic neuron activity in the DLPFC during the delay period of WM tasks (Wilson,
83
O'Scalaidhe et al. 1994), while the injection of GABA antagonist bicuculline disrupts WM
performance in monkeys (Sawaguchi, Matsumura et al. 1989).
Repetitive transcranial magnetic stimulation (rTMS) delivers repeated magnetic pulses to the
cortex to induce plasticity-like changes in cortical function and behaviour. For example, high-
frequency rTMS has been shown to improve language functions in healthy individuals (Sparing,
Mottaghy et al. 2001), and improve different aspects of memory in major depressive disorder
patients (Little, Kimbrell et al. 2000; Martis, Alam et al. 2003; O'Connor, Jerskey et al. 2005).
Additionally, rTMS over the motor cortex has been shown to enhance GABA-mediated
inhibitory neurotransmission in healthy individuals (Daskalakis, Moller et al. 2006). That is,
Daskalakis and colleagues reported a lengthening of the cortical silent period—a measure
reflective of GABAB–mediated inhibitory neurotransmission (Ziemann, Lonnecker et al. 1996)
with increased stimulation frequency, that was maximal at 20 Hz (Daskalakis, Moller et al.
2006). High-frequency (i.e., 20 Hz) rTMS, thus, represents a possible mechanism through which
to potentiate γ oscillatory activity that may, in turn, improve WM performance.
This study, therefore, aimed to evaluate the effect of 20 Hz rTMS applied to the right and left
DLPFC on γ oscillatory activity elicited during the N-back task in healthy subjects. We
hypothesized that active rTMS would enhance γ oscillatory activity with increased WM load
with no effect on oscillatory activity in other frequency ranges (i.e., δ, θ, α, and β).
4.3 Materials and Methods
4.3.1 Subjects
Twenty-two, right-handed healthy volunteers participated in this study (mean age=34.2 years,
SD=7.16 years, range=24-49 years; 11 men and 11 woman). Handedness was confirmed using
the Oldfield Handedness Inventory (Oldfield 1971). All subjects gave their written informed
84
consent and the protocol was approved by the Centre for Addiction and Mental Health in
accordance with the declaration of Helsinki. Exclusion criteria included a self-reported
comorbid medical illness or a history of drug or alcohol abuse. Moreover, psychopathology was
ruled out through the personality assessment screener (PAS; Psychological Assessment
Resources, Inc).
4.3.2 Procedure
Prior to the experiment, subjects were randomized into two groups allocated to receive either
active or sham rTMS. Participants were blind to their group assignment until the completion of
the study to avoid subject biases. The experiment took place over two testing days. On the first
day, subjects performed the N-back test while their EEG activity was recorded. One week later,
rTMS was administered over the DLPFC prior to the final testing in the N-back task. The final
N-back task was performed approximately 20 minutes following the rTMS administration to
allow for cortical plasticity changes to take place as well as for the placement of the EEG cap.
These two N-back testing sessions will be referred to ‗pre‘ and ‗post‘ measures relative to the
rTMS administration here on in.
4.3.3 N-Back Task
Participants performed the N-back task while their EEG activity was recorded (STIM2,
Neuroscan, U.S.A.) prior to (pre) and after (post) a single session of rTMS to the DLPFC.
During this task, stimuli were presented on a computer monitor one at a time to the participants
who were required to push one button (target) if the present stimulus was identical to the
stimulus presented ―N‖ trials back, otherwise the participants pushed a different button (non-
target). Thus, the effect of increasing the cognitive demand on γ oscillatory activity was tested
by varying the ―N‖ of the task in the 0-, 1-, 2- and 3-back conditions (Figure 1). Notably, in the
85
0-back condition, subjects were required to push the non-target button every time a stimulus was
presented, and thus, did not have a memory component. Stimuli consisted of black capital letters
presented for 250 msec followed by a delay period of 3000 msec during which the subject was
required to respond prior to the presentation of a plus sign for 1305 msec indicating the end of
the current trial. In the 0-, 1-, and 2-back condition, stimuli were presented continuously for 15
minutes, while in the 3-back condition; stimuli were presented for 30 minutes. The 3-back
condition was administered for double the length of the other conditions (30 minutes rather than
15 minutes) to ensure that the frequency of target letters in this condition was comparable to
those presented in the 1-, and 2-back conditions, as these target letters occurred less frequently
(i.e., at least 3 letters back). Thus, the proportion of target letters in each of the condition was
23.1%, 15.8%, and 14.8% in the 1-, 2-, and 3-back conditions, respectively. Consequently, this
also ensured that a satisfactory number of correct responses were contained in the all of the
conditions for the data analysis (Table 1). The total time for participants to complete the N-back
task was 1 hour and 15 minutes with the presentation of the conditions randomized and
counterbalanced to prevent order effects.
4.3.4 Repetitive TMS
Repetitive TMS was administered using a Medtronic MagPro stimulator (Medtronic, Inc.,
U.S.A.) with a 70 mm diameter figure-of-8 coil to the right and left DLPFC at 20 Hz, 90 %
resting motor threshold for 25 trains comprising of 30 pulses per train, inter-train interval of 30
seconds for a total of 750 pulses per hemisphere and in accordance with published safety
guidelines (Chen, Gerloff et al. 1997). The time of the rTMS delivery was 25 minutes, 12.5
minutes per hemisphere. The resting motor threshold was defined as the lowest intensity that
produced a motor evoked potential of at least 50 µV in 50% of the trials delivered. Sham
stimulation was delivered with the same rTMS parameters as active stimulation with the coil
86
held in a single wing-tilt position at 90 degrees to induce similar somatic sensations as in the
active stimulation with minimal direct brain effects. The order of stimulation (right then left
versus left then right) was also randomized and counterbalanced to prevent order effects.
4.3.5 DLPFC Site Localization
The localization of the DLPFC was determined through neuronavigational techniques using the
MINIBIRD system (Ascension Technologies) combined with MRIcro/reg software using a T1-
weighted MRI scan obtained for each subject with seven fiducial markers in place. Repetitive
TMS was targeted at the junction of the middle and anterior one-third of the middle frontal gyrus
(Talairach coordinates (x, y, z) = +/- 50, 30, 36) corresponding with posterior regions of
Brodmann area 9 (BA9), and overlapping with the superior region of BA46 (Figure 2). The
selection of this site was based on a recent meta-analysis of functional imaging studies that
examined WM and the activation of the DLPFC (Cannon, Glahn et al. 2005; Mendrek, Kiehl et
al. 2005; Tan, Choo et al. 2005).
4.3.6 EEG Measurement of Evoked γ Oscillatory Activity
Evoked oscillatory responses are phase-locked to the stimulus onset with a fixed latency within
the first 100 msec following stimulus onset and, therefore, can be measured by stimulus-
triggered averaging of responses (Tallon-Baudry, Kreiter et al. 1999). Induced oscillatory
activity, on the other hand, is not phase-locked to stimulus onset and appears as a jitter in latency
that varies from trial to trial, thus, these responses are cancelled out when trials are averaged
(Tallon-Baudry, Kreiter et al. 1999). Both evoked and induced γ activities have been associated
with sensory and cognitive processing (Lutzenberger, Pulvermuller et al. 1995; Muller, Bosch et
al. 1996; Cho, Konecky et al. 2006). Moreover, induced γ oscillatory activity has been shown to
increase with increased cognitive control (Cho, Konecky et al. 2006), while other studies have
87
shown that evoked γ oscillatory activity increases with cognitive demand in WM paradigms
(Howard, Rizzuto et al. 2003; Basar-Eroglu, Brand et al. 2007; Meltzer, Zaveri et al. 2008). As
such, we measured mean evoked γ power from frontal electrodes while subjects completed the
N-back task before (pre) and after (post) rTMS was administered over the right and left DLPFC.
4.3.7 EEG Recording
EEG data were acquired using a 64-electrode cap and Synamps2 DC-coupled EEG system
(Compumedics). Four electrodes placed on the outer side of each eye, above, and below the left
eye were used to monitor eye movement artifact. Data was recorded at a rate of 1000 Hz DC and
with a 0.3 to 200 Hz band pass hardware filter. Electrode impedances were lowered to < 5 kΩ.
All channels were referenced to an electrode placed posterior to the Cz electrode.
4.3.8 Offline EEG processing
To measure the mean evoked γ power, data was filtered off-line using a 1 to 100 Hz band pass
zero phase shift filter (slope, 24 dB/oct). Epochs were defined as –1000 to +3095 msec relative
to the cue onset and were baseline corrected with respect to the prestimulus interval (-1000 to
cue onset). To measure evoked γ power, we selected the entire delay period of 3000 msec
relative to stimulus because the N-back task requires continuous maintenance of ―N‖ letters in
preparation for the next trial as the task runs sequentially for approximately 15-30 minutes
depending on the task condition. All trials were manually inspected and any error trials or
epochs containing artifact (movement or electrooculogram exceeding +/- 50 µV) were excluded
from further analysis.
4.3.9 Data Analysis
Artifact-free EEG data was imported into MATLAB (The MathWorks, Inc. Natick, MA, USA)
using the EEGLAB toolbox (Delorme and Makeig 2004) for subsequent analysis. Evoked γ
88
power in the 30-50 Hz range was averaged over the delay period (0-3000 msec from cue onset)
for the target correct (TC) and non-target correct (NTC) responses for each WM load pre and
post rTMS for each participant. Mean evoked γ power was then assessed during these responses
(TC + NTC) in the frontal electrodes (AF3, AF4, F5, F3, F1, FZ, F2, F4, and F6), and averaged
for each individual. Since spectral analysis of EEG activity is often not normally distributed
(Bender, Schultz et al. 1992), the data was log transformed prior to analysis. Full-factorial
mixed model repeated measures (MMRM) was performed with rTMS (active versus sham) as a
between-subject factor with time (pre versus post) and WM load (0- versus 1- versus 2- versus 3-
back) as within-subject factors on the data with a significance level set at p<0.05. Bonferroni-
adjusted pairwise comparisons were then performed (SAS System v.9.1.3; SAS Institute, NC,
USA).
4.4 Results
4.4.1 Evoked γ Power
As expected, the MMRM analysis revealed an effect of WM load on mean evoked γ power
(F(3,60)=18.59, p0.0001) with subjects generating the greatest γ power during the 2-back
condition compared to the 0- (p<0.0001), 1- (p=0.0416) and 3-back (p=0.0004) conditions,
respectively (Figure 3A). Furthermore, the effect of time on γ power was also found significant
(F(1,20)=29.49, p0.0001) with greater γ power elicited during the N-back following rTMS
stimulation (post) compared to baseline (pre). Finally, the effect of rTMS on γ power was also
significant (F(1,20)=9.26, p=0.0064) with subjects in the active group generating higher γ power
during the N-back task compared to those who received sham stimulation.
The MMRM analysis on mean γ power also revealed significant interaction effects (Figure 3A).
In particular, an interaction was found between time and rTMS (F(1,20) = 12.43, p=0.0021) with
89
active stimulation resulting in greater γ power compared to baseline (pre; p 0.0001) and to sham
rTMS (post; p=0.0028). There was also a significant time x WM load x rTMS interaction:
(F(3,60) = 3.79, p=0.0148). Importantly, while there was no difference in γ power during the N-
back task at baseline (pre) between the active and sham groups, active rTMS resulted in a
significant increase in γ power in the 2- (p=0.0032) and 3-back (p=0.0288) conditions compared
to the sham group. Taken together, these findings suggest that active rTMS resulted in a
significant potentiation in γ power that was greatest in the N-back conditions with the greatest
WM load. Figure 4A shows mean absolute change in γ power as topographical illustrations
(represented by hot colours). Inspection of the topographical plots revealed a maximal change in
γ power in the frontal brain regions compared to other cortical regions following active
stimulation. As such, a repeated measures ANOVA was performed comparing the mean sum
change in γ power in 5 frontal electrodes (FPZ, FP1, FP2, AF3, and AF4) selected based on hot
coloured electrodes from the topographical plot to 5 electrodes from the posterior region (OZ,
O1, O2, PO3, and PO4) with rTMS as a between-subject factor. A significant effect of brain
region was found (F(1,19)=14.938, p 0.001; Figure 4B) and the brain region x rTMS interaction
was also significant (F(1,19)=11.445, p 0.005). Finally, paired t-tests revealed that this
interaction was due to the difference between brain regions following active stimulation (t(10) =
4.101, p 0.005; Figure 4B), while no difference between brain regions was found following
sham stimulation (t(9) = 0.619, p > 0.05; Figure 4B; SPSS 15.0, SPSS Inc. Chicago, Illinois,
USA). These results further suggest that active rTMS resulted in a significant potentiation in γ
power isolated to the frontal brain regions during N-back conditions with the greatest WM load.
90
4.4.2 EEG Spectral Analysis of Other Frequency Bands
To test whether our findings of enhanced mean γ power following active rTMS elicited during
the N-back task was selective to this frequency band, four separate MMRM analyses (active vs.
sham) as a between-subject factor and time (pre vs. post), and WM load (0- versus 1- versus 2-
versus 3-back) as within-subject factors were performed on the log transformed mean evoked δ
(1-3.5 Hz), (4-8 Hz), (9-12 Hz), and (14-28 Hz) power for TC and NTC responses. An
effect of WM load was revealed in all frequency bands; however, there were no differences
between active and sham stimulation under any WM load. Finally, a time x group interaction
was also revealed in the , and frequency bands, but again with no difference between active
and sham frequency power at either baseline (pre) or following rTMS (post). These subsequent
spectral analyses, thus, demonstrate that active stimulation selectively enhanced the oscillatory
activity in the γ band only, while the activity in the other frequencies remained unchanged.
4.4.3 N-Back Working Memory Performance
As expected, the MMRM analysis on WM performance revealed a significant effect of WM load
(F(2, 40)=78.63, p0.001) with performance decreasing with increasing cognitive demand. In
particular, subjects performed significantly better in the 1-back compared to the 2- (p=0.0005)
and 3-back (p<0.0001) conditions and better in the 2-back relative to the 3-back (p<0.0001)
condition, respectively (Table 2). However, the interaction between time and group was not
significant, reflecting similar WM performance across groups (i.e., active versus sham) pre and
post-rTMS (F(1, 20)=0.31, p=0.5859). Finally, no relationship was found between mean WM
performance and mean γ power within any N-back condition determined through a Pearson
correlation coefficient.
91
As expected, we found an effect of WM load on reaction time (F(3,60) = 126.64, p0.0001; Table
3) with reaction time increasing with increasing WM load. Specifically, pairwise comparisons
revealed significantly lower RT in the 0-back relative to the 1-3-back conditions (p0.0001 in
each case), while reaction time was lower in the 1-back compared to the 2- (p=0.0001) and 3-
back condition (p0.0001), respectively.
4.5 Discussion
We measured the effect of 20 Hz rTMS applied to the DLPFC on γ oscillatory activity across
WM load (i.e., 0-, 1-, and 2-back conditions). As predicted, γ oscillatory activity generally
increased with increased WM load. Active rTMS significantly increased γ oscillatory activity
compared to baseline (pre) and sham stimulation. Moreover, active rTMS caused the greatest
change in γ oscillatory activity in the N-back conditions with the greatest cognitive demand, an
effect that was limited to frontal brain regions. Finally, active rTMS had no effect on other
frequency ranges (i.e., δ, , , ) suggesting a selective effect to oscillatory activity in the γ
frequency range. Collectively, these results suggest that active rTMS applied to bilateral DLPFC
significantly increased frontal γ oscillatory activity which was most pronounced at N-back
conditions of greatest difficulty.
The finding that γ oscillatory activity was most pronounced at N-back conditions of greatest
difficulty is consistent with previous studies examining the role of γ oscillatory activity and WM
(Howard, Rizzuto et al. 2003; Basar-Eroglu, Brand et al. 2007; Meltzer, Zaveri et al. 2008). For
example, Howard and colleagues (2003) first demonstrated a linear increase in γ oscillatory
power with WM load in epileptic patients performing the Sternberg WM task (Howard, Rizzuto
et al. 2003). Additional studies confirmed enhanced γ oscillatory activity with WM load
employing both the Sternberg (Meltzer, Zaveri et al. 2008) and N-back (Basar-Eroglu, Brand et
92
al. 2007) tasks. In these studies, however, WM loads comparable to the 3-back condition were
not examined. In this regard, we observed significantly lower γ oscillatory activity in the 3-back
condition relative to the 2-back condition at baseline in both groups (p 0.0004). This finding is
consistent with decreased BOLD activity elicited while healthy subjects performed the 3-back
condition in fMRI studies. It has been suggested that these decreases in BOLD activity may be
because WM resources are exceeded during the 3-back condition resulting in poorer cognitive
performance due to diminished attentional resources (Cowan 2001; Kane, Bleckley et al. 2001;
Wheeler and Treisman 2002). Though speculative, the fact that active rTMS to the DLPFC
resulted in an increase in γ oscillatory activity, with the greatest effect in the 3-back condition
(post; Figure 3A), suggests that rTMS, in part, may enhance attentional resources underlying
such effects.
High-frequency rTMS has been shown to improve cognitive performance (Little, Kimbrell et al.
2000; Sparing, Mottaghy et al. 2001; Martis, Alam et al. 2003; O'Connor, Jerskey et al. 2005)
though the underlying mechanisms have not been investigated. We contend that a possible
explanation is through its effects on GABAergic inhibitory neurotransmission critical in both the
generation and synchronization of oscillatory activity (Whittington, Traub et al. 1995; Wang and
Buzsaki 1996; Bartos, Vida et al. 2007). GABAergic neurons in cortical networks interact with
other neurons to affect spike timing and coordinate rhythmic population activity (Mann and
Paulsen 2007). GABAergic neurons have also been implicated in the synchronization of
pyramidal neurons in the DLPFC during WM (Lewis, Hashimoto et al. 2005). For instance,
Wilson and colleagues (1996) demonstrated that in monkeys, fast spiking GABAergic neurons in
the DLPFC remain active during the delay period of WM tasks (Wilson, O'Scalaidhe et al.
1994). Additionally, during WM tasks, the activity of these GABAergic neurons contribute to
93
the spatial tuning of the neuronal response, and are thus considered to be related to the WM task
itself (Rao, Williams et al. 2000), while the injection of GABAergic antagonists in the DLPFC
resulted in a disruption in WM performance (Sawaguchi, Matsumura et al. 1989). Taken
together with the finding that 20 Hz rTMS potentiates GABAergic inhibitory neurotransmission
(Daskalakis, Moller et al. 2006), we contend that high-frequency rTMS exerts its effects on γ
oscillatory activity via GABA receptor-mediated inhibitory neurotransmission to ultimately
affect WM performance. Future studies combining paired pulse TMS with EEG recording to
index GABA receptor-mediated neurotransmission in the DLPFC (Daskalakis, Farzan et al.
2008) will be needed, however, to determine if γ oscillatory activity is indeed potentiated directly
through enhanced GABA.
Previous studies have also reported modulations in other frequency ranges during cognitive
tasks. Although changes in oscillatory activity in the γ frequency band are most consistently
observed, increases in the , , and frequency bands have also been associated with cognitive
tasks (Gevins, Smith et al. 1997; Klimesch, Doppelmayr et al. 1997; Sarnthein, Petsche et al.
1998; Klimesch 1999; Tallon-Baudry, Kreiter et al. 1999; von Stein, Rappelsberger et al. 1999;
Tesche and Karhu 2000; Klimesch, Doppelmayr et al. 2001; Raghavachari, Kahana et al. 2001;
Jensen, Gelfand et al. 2002; Schack, Vath et al. 2002; Howard, Rizzuto et al. 2003; Rizzuto,
Madsen et al. 2003; Gruber and Muller 2005; Kaiser and Lutzenberger 2005). In line with these
studies, we found an increase in δ, , , and activity with WM load. However, following
rTMS to the DLPFC, modulations were only found within the γ frequency range, while δ, , ,
and activities remained unchanged. These findings indicate that δ, , , , and γ activities are
modulated by cognitive demand yet only γ activity was enhanced through rTMS over the DLPFC
– a brain region that is closely associated with attention and WM performance.
94
To our knowledge, this is the first demonstration of enhanced γ oscillatory activity elicited
during the N-back task following a single session of rTMS over the DLPFC. Although γ
oscillatory activity was potentiated following rTMS, this effect was not related to WM
performance. It may be possible that rTMS induced cognitive changes are either delayed or
optimal at some later time point and/or that repeated sessions (i.e., days to weeks) of rTMS are
needed for such effects to be fully realized. In this experiment we tested subjects in the N-back
task within 20 minutes following rTMS administration, which may have been adequately
captured neurophysiological but not the anticipated cognitive changes. This contention is
consistent with literature from rTMS treatment studies. For example, in patients with major
depressive disorder, high frequency 10 Hz rTMS applied to the left DLPFC two or three times
per week (an average of 10.8 total treatments) was shown to improve cognitive performance and
lessen the number of memory complaints approximately 8.8 days following the last
administration of rTMS (Schultze-Rauchenbach, Harms et al. 2005). Similarly, patients with
Parkinson‘s disease with concurrent depression showed improvements in the Stroop and Harper
& Wisconsin test performances 2- and 8 weeks following 10 sessions of 15 Hz rTMS applied to
the left DLPFC (Boggio, Fregni et al. 2005). It has also been suggested that repeated sessions of
rTMS may be necessary to produce changes in gene expression and synapse formation that
accompany changes in short-term plasticity and cognition (Khedr, Rothwell et al. 2006). Further
studies, therefore, are needed to determine if the anticipated changes to WM performance, that
may follow enhanced γ oscillatory activity, are either delayed or require multiple rTMS sessions
to be fully realized.
Oscillations in the γ frequency range are typically estimated using two sub- classifications that
differ in their phase relationship with respect to the stimulus onset. Evoked γ oscillatory activity
is phase-locked to the stimulus and occurs approximately 200 msec relative to stimulus onset
95
(Tallon-Baudry, Kreiter et al. 1999). Since evoked γ oscillatory activity is phase-locked to the
stimulus, these responses can be extracted in a time domain and averaged across trials (Pantev
1995). In contrast, induced γ oscillatory activity is not phase-locked to the stimulus, but rather
jitters in latency from trial and trial and, thus, cannot be averaged across trials. Induced γ
oscillatory activity is measured by applying the time-frequency decomposition to each trial and
the ensuing power is averaged across trials. The power of evoked and background components
are then subtracted from the total power to provide an estimate of induced γ oscillatory activity
(David, Kilner et al. 2006). Although there are numerous reports that support a relationship
between induced γ oscillatory activity and cognitive functions (Gray, Konig et al. 1989; Fries,
Reynolds et al. 2001; Pesaran, Pezaris et al. 2002; Cho, Konecky et al. 2006), a recent study
demonstrated that measures of induced γ activity may not reflect synchronous neuronal activity,
but rather the activation of miniature saccadic eye movements (Yuval-Greenberg, Tomer et al.
2008). In their study, Yuval-Greenberg et al (2008) recorded eye movements simultaneously
with EEG and found that induced γ activity was time-locked to the onset and the rate of
involuntary miniature saccades and thus, reflects saccadic spike potentials rather than neuronal
oscillations. These authors contend that measures of evoked γ activity may better mitigate such
effects from miniature saccadic activity on this neurophysiological phenomenon (Yuval-
Greenberg, Tomer et al. 2008). Considering that induced γ activity may reflect involuntary eye
movements with emerging evidence supporting the association of evoked γ activity with WM
(Howard, Rizzuto et al. 2003; Basar-Eroglu, Brand et al. 2007; Meltzer, Zaveri et al. 2008), we
selected to measure the effect of rTMS on evoked γ activity.
The results of this study are limited by the relatively small sample size, which may have
attributed to the lack of a relationship found between increased gamma oscillatory activity and
WM performance. Replication studies assessing the effect of rTMS on WM may consider using
96
a larger sample size to further examine this relationship. However, the fact that rTMS enhanced
γ oscillatory activity compared to sham, which was not observed within the other frequency
ranges, suggests that the observed potentiation of γ oscillatory activity was not related to the
small sample size. Nevertheless, such findings should be replicated in a larger sample to
minimize Type II error and stabilize statistical parameter estimates (Norman and Streiner 2000).
A second limitation to this study is the use of the N-back task to evaluate gamma oscillatory
activity underlying WM. Previous studies that have examined the convergent validity between
the N-back task with other WM measures such as the operation span task (OSPAN) show
correlations with lower WM loads (Shelton, Metzger et al. 2007) and varied results with the 3-
back condition (Kane, Conway et al. 2007; Shelton, Metzger et al. 2007). Such studies suggest
that at higher WM loads, attentional components and/or short-term memory processes may be
measured rather than WM capacity exclusively. Our finding of increased gamma oscillatory
activity following rTMS over the DLPFC may, therefore, reflect enhancements in attention
and/or short-term memory processes rather than WM capacity exclusively. Despite these
limitations, however, our findings provide a neurophysiological framework through which to
evaluate rTMS as a therapeutic tool in patient populations (e.g., schizophrenia) where WM
deficits form part of the symptom profile.
In summary, we demonstrated that high-frequency rTMS over the DLPFC significantly enhanced
frontal γ oscillatory elicited in N-back conditions with the greatest task difficulty. Furthermore,
rTMS administered to the DLPFC selectively increased oscillatory activity in the γ frequency
range, while other frequency ranges (i.e., δ, , or ) remained unchanged. The specificity of
high-frequency rTMS on γ oscillatory activity found in this study may, therefore, provide
97
important insight into the pathophysiology of brain disorders that present cognitive deficits as
part of their symptom profiles.
98
Table 1. Mean number of trials (TC + NTC) following artifact correction for each N-back
condition pre and post rTMS administration.
Group Pre
rTMS
Post
rTMS
0-back 1-back 2-back 3-back 0-back 1-back 2-back 3-back
Sham 141.7 127.9 149.7 283.9 125.9 134.3 139.8 264.5
Active 127.9 137.6 124.4 225.8 103.6 129.4 111.7 185.6
Table 2. Performance across N-Back task condition for target and non-target correct responses
expressed as percentage (%) before and after sham or active rTMS. Data expressed as mean (±)
standard deviation (SD).
Group Pre rTMS Post rTMS
1-back 2-back 3-back 1-back 2-back 3-back
Sham
(±) 1SD
92.1 (±) 2.3 88.0 (±) 13.6 77.4 (±) 15.3 94.0 (±) 3.15 89.9 (±) 6.8 80.6 (±) 12.1
Active
(±) 1SD
83.1 (±) 11.6 74.1 (±) 16.3 66.8 (±) 11.5 83.5 (±) 7.8 72.4 (±) 13.3 64.1 (±) 13.9
99
Table 3. Reaction time across N-Back task condition for target and non-target correct responses
expressed in milliseconds (msec) before and after sham or active rTMS. Data expressed as mean
(±) standard deviation (SD).
Group Pre rTMS Post rTMS
0-back 1-back 2-back 3-back 0-back 1-back 2-back 3-back
Sham (±)
1SD
401.4 (±)
110.6
720.8 (+/)
99.3
891.5 (±)
272.7
970.1 (±)
247.7
372.2 (±)
123.9
706.8 (±)
134.2
861.1 (±)
216.5
865.5 (±)
203.1
Active (±)
1SD
393.0 (±)
122.3
793.6 (±)
155.1
924.4 (±)
257.4
990.4 (±)
317.1
435.0 (±)
168.9
756.7 (±)
90.7
862.0 (±)
204.2
956.4 (±)
257.8
100
Figure 1. N-Back WM Task. A) A representation of the four different WM Conditions (0, 1, 2,
and 3-back) that were administered in a randomized order to participants before and after rTMS.
B) The timing of one trial from the presentation of a one letter separated by a (+) sign followed
by a subsequent letter for a total time of 3000 msec. Participants were required to push one
button (target) if the present letter was identical to the letter presented ―N‖ trials back; otherwise
the participants pushed a different button (non-target).
101
Figure 2. Targeting the Dorsolateral Prefrontal Cortex for rTMS Administration. Transverse
view from a single subject with exposed cortex and overlap of Brodmann areas 9 & 46,
highlighted (white) on a T1-weighted 3D MRI. Using MRI-to-MiniBird co-registration, the
centre of the TMS coil was held over this region.
102
Figure 3. Mean log transformed change in oscillatory power (post rTMS frequency power-pre
rTMS frequency power) following rTMS elicited during the N-back task for A) γ power
compared to δ, , , and activities following B) sham, and C) active stimulation. Error bars
represent (+) standard deviation (SD).
103
Figure 4. A) Topographical illustration of mean absolute change (post rTMS γ power-pre rTMS
γ power) in γ power elicited during 3-back condition following sham and active rTMS. Maximal
change in γ power (represented by hot colours) was found following active stimulation in the
frontal brain regions. B) Change in the mean sum of γ power in the frontal
(FPZ+FP1+FP2+AF3+AF4) versus posterior (OZ+O1+O2+PO3+PO4) electrodes in the 3-back
condition following sham and active stimulation. Error bars represent (+) standard deviation
(SD).
104
5 Study 3: The Effect of Transcranial Magnetic Stimulation on Gamma Oscillatory Activity in Schizophrenia
5.1 Abstract
Gamma (γ) oscillations (30-50 Hz) have been shown to be excessive in patients with
schizophrenia (SCZ) during working memory (WM). WM is a cognitive process that involves
the online maintenance and manipulation of information that is mediated largely by the
dorsolateral prefrontal cortex (DLPFC). Repetitive transcranial magnetic stimulation (rTMS)
represents a non-invasive method to stimulate the cortex that has been shown to enhance
cognition and γ oscillatory activity during WM. We examined the effect of 20 Hz rTMS over the
DLPFC on γ oscillatory activity elicited during the N-back task in 24 patients with SCZ
compared to 22 healthy subjects. Prior to rTMS, patients with SCZ elicited excessive γ
oscillatory activity compared to healthy subjects across WM load. Active rTMS resulted in the
inhibition of frontal γ oscillatory activity in patients with SCZ, while potentiating activity in
healthy subjects in the 3-back, the most difficult condition. Further, these effects on γ oscillatory
activity were found to be specific to the frontal brain region and were absent in the posterior
brain region. We suggest that this opposing effect of rTMS on γ oscillatory activity in patients
with SCZ versus healthy subjects may be related to homeostatic plasticity leading to differential
effects of rTMS on γ oscillatory activity depending on baseline differences. These findings
provide important insights into the neurophysiological mechanisms underlying WM and
demonstrated that rTMS can inhibit excessive γ oscillatory activity in SCZ, a finding that may
ultimately translate into a better understanding of the mechanisms leading to cognitive
improvement.
105
5.2 Introduction
Gamma (γ) oscillations (30-50 Hz) are associated with working memory (WM). WM involves
the maintenance and manipulation of information (Baddeley 1986) and has been shown to
increase γ oscillations with increases in WM load in healthy subjects (Basar-Eroglu, Brand et al.
2007), particularly in the dorsolateral prefrontal cortex (DLPFC; (Barr, Farzan et al. 2009)).
Schizophrenia (SCZ) patients have marked deficits in WM (Weinberger, Berman et al. 1986)
that has been attributed to altered γ oscillatory activity. For example, we demonstrated that SCZ
patients compared to healthy subjects elicit excessive γ oscillatory activity while performing the
N-back task at all WM loads that was accompanied by impaired performance (Barr, Farzan et al.
2010). Although previous studies provide evidence for reduced γ oscillatory activity in SCZ
patients during cognitive control (Cho, Konecky et al. 2006) and sensory oddball (Spencer,
Niznikiewicz et al. 2008) tasks, recent findings suggest that during WM performance that γ
oscillatory activity is excessive, even at equivalent performance levels (Basar-Eroglu, Brand et
al. 2007; Barr, Farzan et al. 2010). Altered γ oscillatory activity in SCZ patients may therefore be
related to impaired WM function in this disorder.
Animal studies have shown that γ oscillations during WM are supported by γ –aminobutyric acid
(GABA) interneurons in the DLPFC (Sawaguchi, Matsumura et al. 1989; Wilson, O'Scalaidhe et
al. 1994; Rao, Williams et al. 2000). Specifically, GABAergic activity may be involved in the
generation and inhibition of γ oscillations (Whittington, Traub et al. 1995; Wang and Buzsaki
1996; Bartos, Vida et al. 2007; Brown, Davies et al. 2007), a mechanism that has been shown to
be impaired in SCZ (Benes, McSparren et al. 1991; Akbarian, Kim et al. 1995; Hashimoto, Volk
et al. 2003; Lewis, Hashimoto et al. 2005). In line with these studies, Farzan et al. (2010)
measured neurophysiological indices of GABAB receptor inhibition from the DLPFC in SCZ
106
patients compared to bipolar disordered patients and healthy subjects through combined TMS-
EEG (Farzan, Barr et al. 2010). It was demonstrated that the inhibition of oscillations was
significantly reduced in DLPFC of SCZ patients compared to the other 2 groups (Farzan, Barr et
al. 2010). It is possible, therefore, that deficits in the inhibition of oscillations in the DLPFC in
SCZ results in excessive activity as was previously reported (Barr, Farzan et al. 2010), which
may contribute to WM deficits in this disorder.
By contrast, in healthy subjects it was demonstrated that rTMS over the DLPFC selectively
enhanced γ oscillatory activity that was most pronounced in 3-back (Barr, Farzan et al. 2009),
which may be related to its ability to its potentiating effects on GABAergic neurotransmission
(Daskalakis, Moller et al. 2006; Jung, Shin et al. 2008). The aim of the current study was to
examine the effect of rTMS over the DLPFC on γ oscillatory activity during the N-back task in
SCZ patients compared to healthy subjects. It was hypothesized that rTMS would inhibit
excessive γ oscillatory activity in SCZ patients compared to sham stimulation and in contrast to
healthy subjects.
5.3 Materials and Methods
5.3.1 Subjects
Twenty-four (males=14; females=10) patients with a diagnosis of SCZ or schizoaffective
disorder confirmed by the Structured Clinical Interview for DSM-IV (Spitzer 1994) and 22
(males=11; females=11) healthy individuals participated in this study. All subjects were right
handed confirmed using the Oldfield Handedness Inventory (Oldfield 1971). Patients with SCZ
were all treated with antipsychotic medication (14.4 ± 10.9 mg olanzapine, 6 patients; 233.3 ±
230.9 mg clozapine, 3 patients; 5.2 ± 3.0 mg risperidone, 7 patients; 733.3 ± 416.3 mg of
quetiapine, 3 patients; 2.4 ± 1.6 mg fluphenazine, 3 patients ; 25 mg haloperidol, 1 patient; 15
107
mg aripiprazole, 1 patient). Demographic data of the subject groups are shown in Table 1. The
subject groups were similar in age (t(44)=-0.754, p=0.455), but differed in education (independent
t-tests: t(44)=2.954, p<0.05; Table 1). Severity of psychopathology was evaluated using the
positive and negative symptom scale (PANSS; (Kay, Fiszbein et al. 1987)), scale for the
assessment of negative symptoms (SANS; (Andreasen 1989)) and the Calgary Depression Scale
(CDS; (Addington, Addington et al. 1993); Table 1). Exclusion criteria for all subjects included a
history of substance abuse or dependence in the last 6 months determined through the DSM-IV, a
concomitant major and unstable medical or neurologic illness or pregnant. In healthy subjects the
presence of psychopathology was ruled out through the personality assessment screener (PAS;
Psychological Assessment Resources, Inc). Finally, all subjects provided their written informed
consent and the protocol was approved by the Centre for Addiction and Mental Health in
accordance with the declaration of Helsinki.
5.3.2 Procedure
This study was a randomized, double-blind, placebo-controlled design. Patients with SCZ and
healthy subjects were randomized into two groups allocated to receive either active or sham
rTMS. The experiment took place over two testing days. On the first day, subjects performed the
N-back test while their EEG was recorded. One week later, rTMS was administered over the
DLPFC followed by the final testing of the N-back task. The final N-back task was performed
approximately 20 minutes following the rTMS administration to allow for cortical plasticity to
take place as well as for the placement of the EEG cap. These two N-back testing sessions will
be referred to ‗pre‘ and ‗post‘ measures relative to the rTMS administration here on in.
108
5.3.3 N-Back Task
Subjects performed the N-back task while their EEG was recorded (STIM2, Neuroscan, U.S.A.)
pre and post rTMS. Stimuli were presented on a computer monitor one at a time and participants
were required to push one button (target) if the present stimulus was identical to the stimulus
presented ―N‖ trials back; otherwise, subjects pushed a different button (non-target). Thus, the
effect of increasing cognitive demand on oscillatory activity was tested by varying the ―N‖ in the
1-, 2- and 3-back conditions. Stimuli consisted of black capital letters presented for 250 msec
followed by a delay period of 3000 msec during which the subject was required to respond
(Figure 1). In the 1- and 2-back conditions, stimuli were presented continuously for 15 minutes
and for 30 minutes in the 3-back condition. The 3-back was administered for double the length of
time to ensure a satisfactory number of correct responses were contained for the data analysis
(Table 2). The number of target letters in each condition was: 46 of 198 (23.2%) 1-back; 31 of
197 trials (15.7%) 2-back, and 59 of 400 trials (14.6%) 3-back condition. The N-back task took 1
hour for subjects to complete with the order of conditions randomized and counterbalanced to
prevent order effects.
5.3.4 Repetitive TMS
Repetitive TMS was administered using a Medtronic MagPro stimulator (Medtronic, Inc.,
U.S.A.) with a 70 mm diameter figure-of-8 coil to the right and left DLPFC at 20 Hz, 90 %
resting motor threshold for 25 trains comprising of 30 pulses per train, inter-train interval of 30
seconds for a total of 750 pulses per hemisphere in accordance with published safety guidelines
(Chen, Gerloff et al. 1997). The total time for the rTMS administration was 25 minutes, 12.5
minutes per hemisphere. The resting motor threshold was defined as the lowest intensity that
produced a motor evoked potential of at least 50 µV in 50% of the trials delivered. Sham
stimulation was delivered at the same rTMS parameters as active stimulation with the coil held
109
in a single wing-tilt position at 90 degrees to induce similar somatic sensations as in the active
stimulation with minimal direct brain effects. The order of stimulation (right then left versus left
then right) was also randomized and counterbalanced to prevent order effects.
5.3.5 DLPFC Site Localization
The localization of the DLPFC was determined through neuronavigational techniques using the
MINIBIRD system (Ascension Technologies) combined with MRIcro/reg software using a T1-
weighted MRI scan obtained for each subject with seven fiducial markers in place. Repetitive
TMS was targeted at the junction of the middle and anterior one-third of the middle frontal gyrus
(Talairach coordinates (x, y, z) = +/- 50, 30, 36) corresponding with posterior regions of
Brodmann area 9 (BA9), and overlapping with the superior region of BA46 (Figure 2). The
selection of this site was based on recent meta-analyses of functional imaging studies that
examined WM and the activation of the DLPFC (Cannon, Glahn et al. 2005; Mendrek, Kiehl et
al. 2005; Tan, Choo et al. 2005).
5.3.6 EEG Recording
EEG data were acquired using a 64-electrode cap and Synamps2 DC-coupled EEG system
(Compumedics, U.S.A.). Four electrodes placed on the outer side of each eye, above, and below
the left eye were used to monitor eye movement artifact. Data was recorded at a rate of 1000 Hz
DC and with a 0.3 to 200 Hz band pass hardware filter. Electrode impedances were lowered to <
5 kΩ. All channels were referenced to an electrode placed posterior to the Cz electrode.
5.3.7 Offline EEG processing
We measured mean evoked oscillatory power over the delay period according to published
protocol (Barr, Farzan et al. 2009). Data was filtered off-line using a 1 to 100 Hz band pass zero
phase shift filter (slope, 24 dB/oct). Epochs were defined as –1000 to +3095 msec relative to the
110
cue onset and were baseline corrected with respect to the prestimulus interval (-1000 to cue
onset). All trials were manually inspected and any error trials or epochs containing artifact
(movement or electrooculogram exceeding +/- 50 µV) were excluded from further analysis.
5.3.8 Data Analysis
5.3.8.1 Behavioural Analysis
The total number of correct trials (target correct (TC) and non-target correct (NTC)) including
those trials rejected due to artifact were included in the data analysis for WM performance and
reaction time. Two separate mixed model repeated measures (MMRM) for WM and reaction
were performed were performed on change score (post rTMS-pre rTMS) with Group (patients
with SCZ versus healthy subjects) and rTMS (active versus sham) as between-subject factors and
WM load (1- versus 2- versus 3-back) as the within-subject factor with a significance level set at
p<0.05. Interaction effects were further examined with Bonferroni-adjusted pairwise
comparisons (SAS System v.9.1.3; SAS Institute, NC, USA).
5.3.8.2 EEG Analysis
Artifact-free EEG data were imported into MATLAB (The MathWorks, Inc. Natick, MA, USA)
using the EEGLAB toolbox (Delorme and Makeig 2004) for subsequent analysis. Evoked
oscillatory power for each frequency (δ (1-3.5 Hz); (4-7 Hz); (9-12 Hz); (14-28 Hz) and γ,
(30-50 Hz)) were averaged over the delay period (0-3000 msec from cue onset) for the target
correct (TC) and non-target correct (NTC) responses for each WM load pre and post rTMS for
each subject. Mean evoked oscillatory power was then assessed during these responses (TC and
NTC) from the frontal electrodes (AF3, AF4, F5, F3, F1, FZ, F2, F4, and F6), and averaged for
each subject. Since spectral analysis of EEG activity is often not normally distributed (Bender,
Schultz et al. 1992), the data was log transformed prior to analysis. A series of seven (across
111
oscillatory power in the 5 frequency bands, N-back performance and reaction time) MMRM
were performed on change score (post rTMS-pre rTMS) with Group (patients with SCZ versus
healthy subjects) and rTMS (active versus sham) as between-subject factors and WM load (1-
versus 2- versus 3-back) as the within-subject factor with a significance level set at p<0.05. The
exclusion of the time (pre rTMS versus post rTMS) within subject factor was chosen to simplify
the model to allow for an easier interpretation of a 3-way interaction versus a 4-way interaction
term that would have been highly unstable. As such, the MMRM analyses were carried out on
change scores for oscillatory power and behavioural data (post rTMS-pre rTMS). Interaction
effects were further examined with Bonferroni-adjusted pairwise comparisons (SAS System
v.9.1.3; SAS Institute, NC, USA).
5.4 Results
5.4.1 N-Back Behavioural Performance
Patients with SCZ performed worse than healthy subjects pre and post rTMS, however, there
were no significant improvements observed in N-back performance following either active or
sham rTMS stimulation in both subject groups found through the MMRM analysis (Table 3).
Similarly, the MMRM analysis found no effect of rTMS on reaction time in both subject groups
(Table 3).
5.4.2 Evoked γ Power
Prior to rTMS administration, excessive γ power was observed in patients with SCZ at each WM
load compared to healthy subjects. The MMRM analysis on the change in mean γ power (post
rTMS γ power-pre rTMS γ power) found a significant Group difference between patients with
SCZ and healthy subjects (F(1,42)=18.23; p=0.0001). Further, significant Group x rTMS
(F(1,42)=10.37; p=0.0025) and Group x N-back condition (F(2,42)=6.41; p=0.0037) interaction
112
effects were found. The Group x rTMS x N-back interaction was also significant (F(2,42)=3.75;
p=0.0317; Figure 3A) indicating that the effects of Group and rTMS differed across WM load. A
series of 15 Bonferroni-adjusted pairwise comparisons were then performed to better understand
this 3-way interaction. Active stimulation was found to inhibit γ power in patients with SCZ,
while potentiating γ power in healthy subjects. Moreover, this effect of active stimulation on γ
power differed significantly in patients with SCZ compared to healthy subjects in the 3-back
condition (p<0.0001), while trending differences were observed in the 1- (p=0.0750) and 2-back
(p=0.0795) conditions. To explore whether the effects of rTMS in the 3-back condition were
specific to the frontal brain region, we compared mean γ power from the frontal versus posterior
brain region (electrodes: OZ, O1, O2, PO3, and PO4). A repeated measures ANOVA was
conducted with Group, Time, and rTMS as between subject factors and brain region as a within
subject factor (SPSS 15.0, SPSS Inc. Chicago, Illinois, USA) and a significant Time x Region x
Subject x rTMS (F(1, 35)=9.072; p=0.005; Figure 3B) interaction was revealed. Pairwise
comparisons found no differences in the posterior brain region in both subject groups following
both active and sham stimulation. The effects of active rTMS on γ oscillatory activity therefore
were specific to the frontal brain region in the 3-back condition. These results suggest that active
rTMS over the DLPFC inhibited frontal γ power in patients with SCZ, while potentiating this
activity in healthy subjects that was most pronounced in the 3-back condition with the greatest
difficulty.
5.4.3 EEG Spectral Analysis of Other Frequency Bands
Although γ oscillatory activity is most closely associated with higher order cognitive tasks, we
explored the effect of rTMS on the mean change in oscillatory activities (post rTMS power-pre
rTMS power) in the other frequency bands (δ, , , and ) with four separate MMRM analyses.
113
Although we observed a significant effect of Group on the change in mean oscillatory power in
the , , and frequency bands, no significant Group x rTMS interactions were observed
(Figure 4). However, there was a significant Group x rTMS x N-back interaction found in the δ
frequency band such that active rTMS reduced activity in patients with SCZ in the 3-back
condition compared to sham stimulation (p=0.0048; Figure 4). No differences were found in
healthy subjects in δ activation following rTMS administration.
5.4.4 Effect of Antipsychotic Medication
A Pearson correlation coefficient was performed to determine if the changes in γ and δ
oscillatory activity in the 3-back were related to antipsychotic medication using chloropromazine
equivalents (CPZ; (Woods 2003)) in the SCZ patient group. No relationships were found
between γ or δ oscillatory activities and CPZ equivalents in the 3-back pre or post rTMS
administration.
5.5 Discussion
Our findings suggest that excessive frontal γ oscillatory activity observed in SCZ patients was
significantly inhibited following bilateral rTMS to DLPFC. By contrast, rTMS significantly
potentiated γ oscillatory activity in healthy subjects. These effects were most pronounced in the
3-back and were specific to the frontal cortical regions. rTMS also reduced δ activity in patients
only. These results suggest that rTMS to DLPFC inhibits excessive frontal γ oscillatory activity
during the N-back in SCZ patients an effect that was opposite to that observed in healthy
subjects.
The opposing effect of rTMS on γ oscillatory activity in patients and healthy subjects may be
related to differential changes in GABAergic activity. For example, Daskalakis et al. (2006)
demonstrated that 20 Hz rTMS applied to the motor cortex in healthy subjects had different
114
effects depending on level of baseline GABAergic inhibitory neurotransmission. That is, rTMS
potentiated short interval cortical inhibition (SICI), a neurophysiological paradigm that is related
to GABAA receptor inhibition (Ziemann, Lonnecker et al. 1996), in subjects with relatively low
baseline SICI and suppressed SICI in subjects with relatively high baseline activity (Daskalakis,
Moller et al. 2006) suggesting that rTMS can produce variable effects on GABAA receptor
mediated inhibition depending on baseline levels. As GABAA inhibitory post synaptic potentials
are involved in generation of γ oscillations (Whittington, Traub et al. 1995; Wang and Buzsaki
1996; Bartos, Vida et al. 2007) such findings can be used to explain the variable effects of rTMS
on γ oscillatory activity in SCZ patients and healthy subjects. That is, rTMS inhibited γ
oscillatory activity in SCZ patients with relatively greater γ activity at baseline, while
potentiating activity in healthy subjects with relatively lower γ activity at baseline. Such effects
may also be related to homeostatic plasticity, a brain mechanism that maintains neuronal activity
within a useful physiological range and is critical to neuronal stability (Sejnowski 1977). There
is considerable evidence for the role of GABAA receptor activity in the regulation of homeostatic
plasticity (Hartmann, Bruehl et al. 2008; Le Roux, Amar et al. 2008; Wilhelm and Wenner 2008;
Saliba, Gu et al. 2009; Chen, Shu et al. 2010) by modulating the number of post-synaptic
GABAA receptors that are activated to increase or decrease inhibitory neurotransmission
(Kilman, van Rossum et al. 2002; Swanwick, Murthy et al. 2006; Saliba, Michels et al. 2007;
Hartmann, Bruehl et al. 2008). Regulation of GABAA receptors in homeostatic plasticity have
also been shown to be involved in the synchronization of neuronal activity (Galarreta and Hestrin
1999; Gibson, Beierlein et al. 1999; Koos and Tepper 1999; Tamas, Buhl et al. 2000). The
opposing effect of rTMS on γ oscillatory activity in the current study, therefore, may reflect
differential regulation of inhibitory activity through efficacy of GABAA receptors important in
homeostatic plasticity and generation of γ oscillations.
115
Alternatively, the effect of rTMS on γ oscillatory activity may reflect the regulation of cortical
excitability to maintain homeostatic plasticity as GABA neurons are dependent on excitatory
drive in generation of γ oscillations (Whittington, Traub et al. 1995; Traub, Whittington et al.
1996; Traub, Michelson-Law et al. 2004; Bartos, Vida et al. 2007; Mann and Paulsen 2007). In
this regard, the main source of neuronal excitation is through release of glutamate which
typically activates N-methyl-D-asparate (NMDA) and non-NMDA receptors in the post-synaptic
membrane (Gonzalez-Burgos and Lewis 2008). The duration of non-NMDA excitatory post
synaptic potentials (EPSPs) are optimal for fast signaling and coincidence detection. As such,
non-NMDA EPSPs are important in the precise control of spike timing needed in the
synchronization of cortical oscillations. The generation of oscillatory activity, therefore, may not
only depend on GABA mediated inhibition but also on the recruitment of interneuron firing by
glutamate excitation (Gonzalez-Burgos and Lewis 2008). Homeostatic plasticity has been shown
through the alteration of cortical excitability with transcranial direct current stimulation (tDCS)
and rTMS administered to the motor cortex in healthy subjects (Siebner, Lang et al. 2004). In
their study, cathodal tDCS reduced corticospinal excitability followed by 1 Hz rTMS that
resulted in a sustained increase in corticospinal excitability. By contrast, increased corticospinal
excitability by anodal tDCS was subsequently reduced with 1 Hz rTMS. Those subjects with the
greatest changes induced by tDCS priming also exhibited the greatest change in corticospinal
excitability following rTMS (Siebner, Lang et al. 2004). Siebner et al. (2004) therefore
demonstrate that rTMS can produce variable effects on cortical excitability depending on
baseline activity level. Given the importance of excitatory drive on γ oscillations, the
homeostatic regulation of cortical excitability through rTMS may also have produced our finding
of opposing effects on γ oscillatory in patients versus healthy subjects.
116
As previously shown, rTMS had no effect on the δ, , , and frequency bands in healthy
subjects (Barr, Farzan et al. 2009). In SCZ patients, however, rTMS reduced δ oscillatory
activity in the 3-back compared to sham stimulation. This finding is consistent with a previous
rTMS study in patients with SCZ with predominant negative symptoms (Jandl, Bittner et al.
2005). In this study, when rTMS was applied at 10 Hz to the left DLPFC for 5 days a reduction
in negative symptoms and δ oscillatory activity was reported (Jandl, Bittner et al. 2005). It is
possible that a reduced δ activity following rTMS in SCZ patients may be related to inhibition of
γ oscillatory activity given the non-random relationships between oscillatory frequencies
(Roopun, Kramer et al. 2008). That is, cross-frequency interactions or ―nesting‖ is observed
when the power of a discrete frequency band is modified by the phase of a lower frequency band
that coexists during information processing (Roopun, Kramer et al. 2008). In this regard,
hierarchies of nested rhythms have been observed in the neocortex (Penttonen and Buzsaki 2003)
between δ, , and γ oscillatory activities. These findings suggest that γ oscillations are nested
within δ oscillations and in relation to our findings, the effect of rTMS on γ activity in SCZ
patients may, in part, be related to an altered δ oscillatory activity.
This study is limited in some important ways. First, although rTMS inhibited excessive γ
oscillatory activity in SCZ patients and potentiated activity in healthy subjects, this change was
not related to improved WM performance. Previous studies, however, have shown that rTMS
induced cognitive changes are either delayed or optimal at some later time point (Boggio, Fregni
et al. 2005; Schulze-Rauschenbach, Harms et al. 2005) and that repeated rTMS sessions may be
needed to produce changes in gene expression and synapse formation associated with changes in
short-term plasticity and cognition (Khedr, Rothwell et al. 2006). Nevertheless, this study
provides early and interesting neurophysiological evidence for the modulating effect of rTMS on
117
γ oscillatory activity, a finding that warrants further investigation as a potential therapeutic
mechanism which underlies WM impairments in SCZ. Second, the relatively small sample size
tested may be insufficient to detect an improvement in performance on the N-back following
rTMS. Replication studies may consider using a larger sample size to examine this relationship.
However, the fact that rTMS altered γ oscillatory activity compared to sham, suggests that the
change in γ was not related to the small sample size. Nevertheless, such findings should be
replicated in a larger sample to minimize Type II error and stabilize statistical parameter
estimates (Norman and Streiner 2000). Third, our finding of reduced γ oscillatory activity
following rTMS may be related to the effect of antipsychotic medication. In this regard, Hong et
al. (2004) reported enhanced 40 Hz oscillations in SCZ patients that were treated with second
generation compared those taking conventional antipsychotics (Hong, Summerfelt et al. 2004).
In this study, oscillations were parsed into 20, 30 and 40 Hz; however, 30-50 Hz range is most
conventionally examined during cognitive tasks (Howard, Rizzuto et al. 2003). Similarly, a
differential effect of antipsychotic medication on cognitive performance has also been reported
with SCZ patients on second generation performing better than those patients on conventional
antipsychotics on a variety of cognitive tests, including WM (Meltzer and McGurk 1999; Bilder,
Goldman et al. 2002; Sharma, Hughes et al. 2003). In our sample, only 4 subjects were on
conventional antipsychotics and these subjects happened to be randomly assigned to the sham
group. Nevertheless, there were no differences found in γ oscillatory activity or in WM
performance pre or post sham rTMS in those patients on conventional versus second generation
antipsychotics. We were unable therefore to evaluate the effect of rTMS on γ oscillatory activity
in SCZ patients on conventional versus second generation antipsychotics.
In summary, we demonstrated that rTMS over DLPFC alters frontal γ oscillatory activity with
the greatest effect at higher WM loads. In patients, rTMS inhibited excessive γ oscillatory
118
activity across WM load. In contrast, rTMS potentiated γ oscillatory activity in healthy subjects.
The differential effect of rTMS on γ oscillatory activity may be related to the concept of
homeostatic plasticity involving the regulation of GABAergic inhibitory mechanisms that
maintain neuronal excitability within a useful physiological range. These findings provide
important insights into the neurophysiological mechanisms that may lead to cognitive
potentiation in this disorder.
119
Table 1. Demographic Data for Healthy Subjects (HS) and patients with schizophrenia (SCZ)
and the assessment of psychotic symptoms in patients with SCZ rTMS (±) 1 standard deviation.
HS SCZ
Age 44.500 (±) 11.43 47.21 (±) 12.80
Age Range 23-61 23-70
Female (n) 11 10
Male (n) 11 14
PANSS Scores Positive NA 17.50 (±) 6.40
Negative NA 14.71 (±) 7.09
Global NA 27.45 (±) 7.60
Total NA 57.54 (±) 16.05
Psyrats Score Total NA 13.42 (±) 13.52
CDS Score Total NA 2.94 (±) 2.99
SANS Score Total NA 37.67 (±) 21.42
120
Table 2. Total number (TC+NTC) of trials analyzed for healthy subjects (HS) and patients with
schizophrenia (SCZ) in the 1-, 2-, and 3-back task conditions pre- post-rTMS (±) 1 standard
deviation.
No of Trials Condition HS SCZ
Pre Post Pre Post
Active Sham Active Sham Active Sham Active Sham
1-Back 130.91 130.00 127.09 114.81 82.00 84.17 104.00 81.00
(±) SD 39.77 24.61 42.37 48.82 42.21 43.79 35.60 41.97
2-Back 150.00 130.36 139.82 104.91 101.92 96.08 54.42 90.50
(±) SD 32.75 33.14 43.21 26.36 42.39 44.07 36.39 45.57
3-Back 283.82 225.64 264.45 181.64 157.27 168.42 122.08 131.83
(±) SD 76.22 62.92 79.65 77.80 73.38 76.71 59.42 78.26
121
Table 3. Working memory (WM) behavioural performance (%) and reaction time (RT; msec)
during the N-back in healthy subjects (HS) versus patients with schizophrenia (SCZ) pre-post
either active or sham rTMS (±) 1 standard deviation pre-post rTMS.
Behavioural Condition HS SCZ
Pre Post Pre Post
Active Sham Active Sham Active Sham Active Sham
Score (%) 1-Back 83.19 92.12 83.58 94.01 79.05 77.33 76.22 69.42
(±) SD 11.61 2.26 7.80 3.15 15.16 18.31 21.07 29.75
2-Back 74.18 88.03 72.49 89.93 66.25 72.46 62.95 79.61
(±) SD 16.29 13.57 13.32 6.87 11.58 22.99 24.11 16.72
3-Back 66.86 77.41 60.85 62.65 48.38 59.99 48.55 57.46
(±) SD 11.58 15.39 3.12 3.71 16.08 20.84 15.40 15.97
RT (msec) 1-Back 793.56 720.78 756.73 706.81 945.02 754.36 874.35 796.48
(±) SD 147.18 99.26 90.68 134.15 276.11 192.41 277.71 192.70
2-Back 924.42 891.49 861.96 861.11 1130.73 865.01 1012.17 902.22
(±) SD 257.44 272.74 204.21 216.49 434.34 245.49 340.66 288.48
3-Back 990.42 970.08 956.42 865.53 1114.51 938.66 937.35 929.60
(±) SD 317.13 247.70 257.75 203.11 277.71 236.39 322.09 252.02
122
Figure 1. A) A representation of the 1-, 2- and 3-back conditions that were completed in a
randomized order by patients with schizophrenia (SCZ) and healthy subjects (HS) pre-post
rTMS. Subjects were required to push one button (target) if the current letter was identical to the
letter presented ―N‖ trials back; otherwise the participants pushed a different button (non-target).
Correct responses for target (TC) and non-target (NTC) were included in the data analysis. B)
The timing of one trial from the presentation of a one letter separated by a (+) sign followed by a
subsequent letter for a total time of 3000 msec.
123
Figure 2. Targeting the Dorsolateral Prefrontal Cortex (DLPFC) for rTMS stimulation.
Transverse view from a single subject with exposed cortex and overlap of Brodmann areas 9 &
46, highlighted (white) on a T1-weighted 3D MRI. Using MRI-to-MiniBird co-registration, the
centre of the TMS coil was held over this region.
124
Figure 3. A) Mean log transformed gamma oscillatory power (γ; 30-50 Hz; uV2) for target
correct (TC) and non-target correct (NTC) responses during the N-back task pre-post rTMS in
healthy subjects (HS; N=22) versus patients with schizophrenia (SCZ; N=24). B) Mean log
transformed γ oscillatory power (uV2) for target correct (TC) and non-target correct (NTC)
responses during the 3-back condition measured from the frontal and posterior brain regions pre-
post rTMS in patients with schizophrenia (SCZ; N=24) and healthy subjects (HS; N=22). Bars
represent (±) 1 standard deviation.
125
Figure 4. Mean log transformed oscillatory power (uV2) for target correct (TC) and non-target
correct (NTC) responses during the N-back task pre-post rTMS in healthy subjects (HS; N=22)
versus patients with schizophrenia (SCZ; N=24) across delta (δ; 1-3.5 Hz), theta (; 4-7 Hz),
alpha (; 8-12 Hz), and beta (; 12.5-28 Hz) frequency ranges. Bars represent (±) 1 standard
deviation.
126
6 Discussion
6.1 Summary of Results
The results presented in the previous chapters consist of three main findings: (a) patients with
schizophrenia compared to healthy subjects generate excessive frontal evoked gamma oscillatory
activity during the N-back task regardless of working memory load and at equivalent
performance levels; (b) in healthy subjects, rTMS applied to the DLPFC selectively potentiates
frontal evoked gamma oscillatory activity during a subsequent testing on the N-back task; and (c)
rTMS over the DLPFC inhibits excessive frontal evoked gamma oscillatory in patients with
schizophrenia in contrast to healthy subjects.
Study one measured frontal evoked oscillatory activity while performing the N-back working
memory task in patients with schizophrenia compared to healthy subjects based on our a priori
hypothesis that patients would exhibit an alteration in the gamma frequency band. We
demonstrated two main group differences. First, patients with schizophrenia generated excessive
frontal evoked gamma oscillatory activity compared to healthy subjects regardless of working
memory load and at equivalent performance levels. Second, we observed reduced frontal evoked
beta oscillatory at each working memory load in patients with schizophrenia compared to the
healthy subject group. Furthermore, this study explored the relationship between frontal evoked
oscillatory activity generated during the 3-back, working memory performance and symptom
severity. We observed an inverse relationship between working memory performance and the
severity of negative but not positive symptoms. The findings of this study, therefore, demonstrate
that patients with schizophrenia exhibit impaired frontal evoked gamma oscillatory activity while
performing the N-back task and raised two important questions. First, can high frequency rTMS
applied to the DLPFC normalize excessive frontal evoked gamma oscillatory activity generated
127
during the N-back task in patients with schizophrenia? Second, if rTMS modulates frontal
evoked gamma oscillatory activity, would these neurophysiological changes be related to
improved performance on the N-back task? To answer these two questions, we administered high
frequency rTMS to the DLPFC first in healthy subjects in study two and patients with
schizophrenia in study three.
In the second study, we administered one session of 20 Hz rTMS applied bilaterally to the
DLPFC to evaluate its effects on frontal evoked gamma oscillatory activity in healthy subjects on
a subsequent testing on the N-back task. We hypothesized that rTMS would enhance frontal
evoked gamma oscillatory activity based on previous studies which have shown that rTMS
enhances oscillatory activity (Okamura, Jing et al. 2001; Griskova, Ruksenas et al. 2007) and
GABAergic activity (Daskalakis, Moller et al. 2006) which have been shown to underlie the
generation and modulation of gamma oscillations (Whittington, Traub et al. 1995; Wang and
Buzsaki 1996; Bartos, Vida et al. 2007). It was shown that rTMS applied to the DLPFC
potentiated frontal evoked gamma oscillatory that was most pronounced in the more difficult
working memory loads (i.e., 2- and 3-back conditions) compared to baseline activity and sham
stimulation. Furthermore, we showed that this effect was isolated to the frontal brain region and
was specific to activity in the gamma frequency range. This study demonstrated the ability of
rTMS applied over the DLPFC to selectively potentiate frontal evoked gamma oscillatory
activity in healthy subjects and raised one important question. Can excessive activity in patients
with schizophrenia observed in study 1 be normalized by rTMS and improve working memory
performance?
Study three administered 20 Hz rTMS applied bilaterally to the DLPFC to measure its effect on
frontal evoked gamma oscillatory activity elicited during the N-back task in patients with
128
schizophrenia compared to healthy subjects. It was hypothesized that rTMS would inhibit
excessive frontal evoked gamma oscillations in patients with schizophrenia compared to sham
stimulation and in contrast to healthy subjects. It was demonstrated that rTMS over the DLPFC
resulted in the inhibition of frontal evoked gamma oscillatory activity that was most pronounced
in the 3-back condition. By contrast, rTMS potentiated frontal evoked gamma oscillatory activity
in healthy subjects thereby replicating our finding from study two. We suggested that the
differential effect of rTMS on frontal gamma oscillatory activity was due to differences in
baseline activity levels in patients with schizophrenia compared to healthy subjects, a concept
that may be related to homeostatic plasticity. Such homeostatic effects may be mediated through
GABAergic inhibitory neurotransmission given its important role in the both the generation and
modulation of gamma oscillatory activity.
6.2 The Effect of rTMS on Frontal Evoked Gamma Oscillatory Activity and the Concept of Homeostatic Plasticity
The role of GABAergic inhibitory neurotransmission in the generation and modulation of
gamma oscillations has been emphasized in this thesis. However the excitatory drive mediated
by glutamate to GABA neurons has also been shown to be critical in the generation of gamma
oscillations (Whittington, Traub et al. 1995; Traub, Whittington et al. 1996; Traub, Michelson-
Law et al. 2004; Bartos, Vida et al. 2007; Mann and Paulsen 2007). The balance or homeostasis
between excitatory and inhibitory neurotransmission therefore is important in the regulation of
gamma oscillations. Unlike slow-onset homeostasis that is involved in the adaptation to one‘s
environment, fast-onset homeostatic regulation is needed to quickly counteract any
destabilization in function of the neuronal network (Chen, Chen et al. 2008). That is, the
functional state of most central neuronal networks is highly dynamic which causes rapid changes
in neuronal firing patterns, overall activity, and synchrony (Buzsaki and Draguhn 2004). In
129
regards to cortical oscillations, the pattern of neuronal activity depends not only on the
architecture of the neuron, but also on its initial conditions (Glass 2001; Penttonen and Buzsaki
2003). That is, unless an oscillator is perturbed, the pattern of neuronal activity will repeat
indefinitely in a noise-free system (Lisman and Idiart 1995). In relation to the findings of study
three, the effect of rTMS on gamma oscillatory activity may have perturbed these oscillations
which may have resulted in homeostatic mechanisms to regulate this activity based on initial
baseline levels. That is, in patients with schizophrenia with initial excessive gamma oscillatory
activity, rTMS resulted in the inhibition of this activity while potentiating this activity in healthy
subjects with relatively lower baseline levels. This suggestion is consistent with previous rTMS
studies. For example, rTMS has been shown to produce variable effects on inhibitory (Bagnato,
Curra et al. 2005; Daskalakis, Moller et al. 2006) and excitatory (Siebner, Lang et al. 2004)
mechanisms that were dependent on the level of baseline activity thereby demonstrating
homeostatic mechanisms through rTMS.
Homeostatic plasticity is an important brain mechanism that maintains neuronal activity within a
useful physiological range and is critical to neuronal stability and survival (Sejnowski 1977). For
proper information processing, the level of activity in neuronal networks has to be maintained
within a physiological range between absolute silence and over-excitation that occurs during
seizures (Hartmann, Bruehl et al. 2008). In patients with schizophrenia, excessive gamma
oscillatory activity was accompanied by deficits in working memory performance. Such
excessive activity may reflect a dsyregulation of homeostatic mechanisms as neural plasticity has
been shown to be altered in patients with schizophrenia (Daskalakis, Christensen et al. 2008).
Although, the administration of rTMS resulted in the inhibition of these oscillations, this
neurophysiological change was not associated with improved working memory performance.
Furthermore, oscillatory activity was measured within 20 minutes following rTMS stimulation
130
and it is possible that gamma oscillatory activity may have returned to baseline levels following
the testing session. In this regard, it has been suggested that repeated sessions of rTMS are
needed produce changes in gene expression and synapse formation associated with changes in
short-term plasticity and cognition (Khedr, Rothwell et al. 2006). By extension, repeated
administration of rTMS may produce long lasting changes in oscillatory activity that may
ultimately translate into improved working memory performance. Nevertheless, study three
demonstrated that one session of rTMS may have perturbed gamma oscillatory activity resulting
in variable effects on this activity in patients with schizophrenia and healthy subjects based on
baseline activity. This finding may therefore reflect homeostatic mechanisms in the regulation of
the excitatory and inhibitory balance important in gamma oscillations.
6.3 Is there an Optimal Level of Gamma Oscillatory Activity in Working Memory?
In the previous section, it was suggested that rTMS may serve to maintain frontal evoked gamma
oscillatory activity within a useful physiological range. As described in chapter 1, oscillatory
activity may be involved in the temporal framework for information encoding (Singer 1999;
Fries, Nikolic et al. 2007) and that the interaction between activities in the various frequency
bands may provide the brain with multiple times scales (Buzsaki 2006). In regards to working
memory, it has been shown that gamma activity increases in response to increases in working
memory load (Howard, Rizzuto et al. 2003) which has led to the hypothesis that gamma activity
is used to organize and temporally segment the representations of different items in a multi-item
working memory system (Lisman and Idiart 1995; Tallon-Baudry, Bertrand et al. 1996; Luck
and Vogel 1997; Jensen and Lisman 1998; Pesaran, Pezaris et al. 2002). In line with this
hypothesis, the series of studies presented in this thesis demonstrated that frontal evoked gamma
oscillatory activity in the DLPFC increases in response to working memory load from the 0- to
131
the 2-back condition followed by a decrease in activity in the 3-back condition in healthy
subjects resulting in an inverted U-shaped pattern. This inverted U-shaped pattern has also been
reflected by the BOLD signal measured in an fMRI studies in healthy subjects performing the N-
back task (Callicott, Mattay et al. 1999) that has been suggested to result from diminished
attentional resources in the 3-back condition (Cowan 2001; Kane, Bleckley et al. 2001; Wheeler
and Treisman 2002). The increase in frontal evoked gamma oscillatory activity following rTMS
in healthy subjects shown in studies two and three may have been attributed to increased
attention that is a critical component of working memory particularly during higher working
memory loads (Shelton, Metzger et al. 2007). In this regard, attention has also been shown to
modulate with gamma oscillatory activity (Brovelli, Lachaux et al. 2005; Vidal, Chaumon et al.
2006) and may reflect what aspect of working memory that was potentiated through rTMS
resulting in a more linear pattern of frontal evoked gamma oscillatory activity in response to
working memory load. In healthy subjects, therefore, increased gamma oscillatory activity
appears to be beneficial while performing working memory tasks especially during higher
working memory loads resulting in a more linear pattern of gamma activity in response to
working memory load.
In patients with schizophrenia, the findings of studies one and three revealed that patients with
schizophrenia generated excessive and similar frontal evoked gamma oscillatory activity across
working memory load in contrast to the inverted U-shaped pattern observed in healthy subjects.
Heightened levels of activity regardless of task demand has also been shown by Basar-Erolgu et
al 2007 with EEG (Basar-Eroglu, Brand et al. 2007) and also with fMRI (Callicott, Mattay et al.
2003). Although from previous studies and the healthy subject findings presented in this thesis, it
is hypothesized that greater gamma activity in response to working memory load is
advantageous; however, too much activity does not optimize performance on these tasks at least
132
in patients with schizophrenia. This raises the following questions: is there an optimal level of
gamma for working memory performance and is this level of gamma different for patients with
schizophrenia and healthy subjects? The fact that increased frontal evoked gamma oscillatory
activity was accompanied by impaired working memory performance in patients with
schizophrenia compared to healthy subjects observed in studies one and three suggests that too
much gamma is not advantageous in this disorder. In healthy subjects, however, study three
showed that rTMS potentiated frontal evoked gamma oscillatory to a similar baseline level of
gamma in patients with schizophrenia in the 3-back condition. It is therefore possible that rTMS
may have had a deleterious effect on gamma activity in healthy subjects in generating too much
gamma however we are unable to evaluate this possibility as there was no change in behavioural
performance in healthy subjects following rTMS. Future studies are needed to investigate if there
is an optimal level of gamma activity needed for working memory and if this level of gamma is
different in patients with schizophrenia compared to healthy subjects. Alternatively, it is possible
that the modulation of gamma in response to working memory load is critical to optimal working
memory function rather than the amount of gamma generated. This possibility will be discussed
in the following section.
6.4 The Modulation of Gamma Activity in Working Memory Through Cortical Inhibition
The results from studies one and three demonstrated that patients with schizophrenia fail to
modulate frontal evoked gamma activity in response to increased working memory load. This
finding is consistent with previous EEG studies testing working memory (Basar-Eroglu, Brand et
al. 2007) and cognitive control (Cho, Konecky et al. 2006) and also with fMRI (Snitz,
MacDonald et al. 2005). Taken together, these studies suggest that the lack of modulating
gamma activity in response to working memory load may contribute to working memory deficits
133
in patients with schizophrenia. As previously described, cortical inhibition plays a critical role in
gamma oscillations. Specifically is has been shown that GABAA receptor mediated IPSPs are
involved in the generation of gamma oscillations (Whittington, Traub et al. 1995; Wang and
Buzsaki 1996; Bartos, Vida et al. 2007), while GABAB receptor activity is involved in the
modulation of gamma oscillations (Whittington, Traub et al. 1995; Brown, Davies et al. 2007;
Leung and Shen 2007). In studies one and three, we demonstrated that patients with
schizophrenia elicit excessive frontal evoked gamma oscillatory activity compared to healthy
subjects regardless of working memory load and at equivalent performance levels. This finding
suggests that in patients with schizophrenia there is impairment in the modulation of gamma
oscillatory activity mediated by GABAB receptor activity rather than a deficit in its generation.
In this regard, our lab has recently measured neurophysiological indexes of GABAB receptor
mediated inhibition from the DLPFC in patients with schizophrenia compared to patients with
bipolar disorder and healthy subjects through combined TMS-EEG (Farzan, Barr et al. 2010). It
was demonstrated that the inhibition of gamma oscillations were significantly reduced in the
DLPFC in patients with schizophrenia compared to the other two groups (Farzan, Barr et al.
2010). It is therefore possible that the lack of inhibition of gamma oscillations in schizophrenia
may be related to our finding of excessive frontal evoked gamma oscillatory activity generated
during the N-back task observed in study one and prior to rTMS administration in study three. It
follows that the inhibition of frontal evoked gamma oscillatory activity following rTMS in
patients with schizophrenia may have enhanced the activity of GABAB receptor mediated
inhibitory neurotransmission. In this regard, Daskalakis et al (2006) demonstrated that the
application of a single session of 20 Hz rTMS to the motor cortex resulted in the significant
enhancement of a TMS neurophysiological index of GABAB receptor mediated inhibition in
healthy subjects (Daskalakis, Moller et al. 2006). Together these studies suggest that the lack of
134
gamma modulation in patients may contribute to the deficits observed in working memory and
that rTMS may have exerted its effect on frontal evoked gamma oscillatory activity through
GABAergic inhibitory mechanisms.
6.5 Medication Effects on Cortical Excitability and Oscillatory Activity
As previously discussed, gamma oscillations are dependent on excitatory and inhibitory
neurotransmission. Further, it has been demonstrated that antipsychotic medications influence
excitatory and inhibitory activity in patients with schizophrenia however these findings have
produced inconsistent findings. For example, Pascual-Leone et al (2002) reported increased
motor threshold and decreased cortical inhibition in patients with schizophrenia on conventional
antipsychotic medication, while unmedicated patients did not exhibit such differences compared
to healthy subjects (Pascual-Leone, Manoach et al. 2002). By contrast, Daskalakis et al (2002)
reported significant cortical inhibition deficits in unmedicated patients with schizophrenia,
whereas no differences were found between medicated (conventional and unconventional
antipsychotics) patients compared to healthy subjects (Daskalakis, Christensen et al. 2002).
Despite these divergent findings, deficits in cortical inhibition is a robust finding in patients with
schizophrenia and has been suggested to result from: (1) excessive subcortical dopaminergic
activity that results in either the disinhibition of cortical inhibitory neurotransmission (Walker
1994) or decreased activation of cortical inhibitory projections (Swerdlow and Koob 1987); and
(2) reduced GABAergic interneurons in the prefrontal, anterior cingulate, and hippocampal
formation (Benes, McSparren et al. 1991; Benes 1999). It has further been demonstrated that
unconventional antipsychotic medications have different effects on cortical inhibition. For
example, three studies conducted by the Daskalakis group indicate differences in cortical
inhibition between patients treated with clozapine, olanzapine, and risperidone (Fitzgerald,
135
Brown et al. 2002; Daskalakis, Christensen et al. 2008; Liu, Fitzgerald et al. 2009). It was
suggested that such medications may differ in their action on GABA, glutamate or through the
modulation of ascending amine systems. The interaction of ascending dopaminergic and
serotonergic pathways and cortical networks with both excitatory and inhibitory cortical
connections is however still not fully understood. Taken together, these studies suggest that
medication influences both cortical excitatory and inhibitory mechanisms in patients with
schizophrenia through the interaction of several neurotransmitters which may in turn result in
differential effects on gamma oscillatory activity.
As previously described, GABA plays a critical role in the generation and modulation of gamma
oscillations. Antipsychotic medication that exerts its influence on GABAergic inhibitory
neurotransmission may therefore differentially effect gamma oscillations compared to other
antipsychotics. For example, clozapine may result in the blockade of GABAA (Squires and
Saederup 2000) and has been shown to increase GABAB receptor activity (Daskalakis,
Christensen et al. 2008) may modulate gamma oscillations in patients with schizophrenia.
Alternatively, the blockade of GABAA receptors by clozapine has been shown to increase the
amplitude of theta and gamma oscillations (Squires and Saederup 2000) possibly through basket
cells which can synchronize these frequencies in pyramidal cells by inducing rebound action
potentials following GABAergic hyperpolarizing IPSPs (Cobb, Buhl et al. 1995). In addition,
when GABA binds to GABAB receptors, the release of dopamine, noradrenergic, and serotonin
are inhibited (von Bohlen 2006) thus the action of clozapine on these other neurotransmitter
systems may also influence gamma oscillations in patients with schizophrenia. In this regard,
clozapine also interrupts the binding of dopamine at D1, D2, D3, and D5 receptors (Naheed and
Green 2001), which may have resulted in excessive gamma oscillatory activity. For example, it
has been shown in non-human primates that the blockade of D1 and D5 receptors during a
136
working memory task enhances task-related excitability (Williams and Goldman-Rakic 1995).
This finding therefore suggests that clozapine‘s antagonism of dopamine receptors may result in
the excessive gamma oscillatory activity observed in patients with schizophrenia in studies one
and three.
The effect of antipsychotic medications on lower frequencies in schizophrenia animal models has
also been examined. For example, alteration of low frequency oscillatory activity by 5-HT2A/2C
agonist DOI and noncompetitive NMDA-R antagonist phencyclidine (PCP) have been shown to
reduce activity in the 3-4 Hz range that was subsequently reversed by application of both
haloperidol and clozapine (Kargieman, Santana et al. 2007; Celada, Puig et al. 2008). Another
study which examined the effect of acute haloperidol and chronic risperidone treatment on
resting, evoked and induced oscillatory activities in a mice schizophrenia model reported that
antipsychotic medications reduced resting theta and increased evoked gamma oscillatory activity
during an auditory task. Similar to clozapine, therefore, haloperidol and risperidone treatment
may have contributed to excessive gamma oscillatory activity in patients with schizophrenia
demonstrated in studies one and three.
Taken together, these studies reveal differential effects of antipsychotic medications on cortical
inhibition in patients with schizophrenia and on oscillatory activities with the use of
schizophrenia animal models. Such development of animal models with altered EEG spectra can
evaluate the contributions of GABA, dopaminergic and glutamatergic states in schizophrenia and
also evaluate the confounding effects of antipsychotic medication on human EEG studies.
Although an effect of medication was not found in studies one and three, these studies reviewed
above suggest that antipsychotic medication that block GABAA or dopamine receptors may
contribute to the excessive gamma oscillatory activity found in patients with schizophrenia.
137
6.6 Hypo or Hyperfunctioning of the Prefrontal Cortex Contribute to Cognitive Deficits?
There is considerable evidence for abnormal functioning of the prefrontal cortex associated with
working memory deficits in patients with schizophrenia. The nature of these deficits, however,
remains controversial. For example, fMRI studies have demonstrated both enhanced (Callicott,
Bertolino et al. 2000; Manoach, Gollub et al. 2000), reduced (Callicott, Ramsey et al. 1998;
Barch, Carter et al. 2001; Perlstein, Carter et al. 2001; Barch, Csernansky et al. 2002) or no
difference (Honey, Bullmore et al. 2002; Walter, Wunderlich et al. 2003; Kindermann, Brown et
al. 2004; Walter, Vasic et al. 2007) in DLPFC activation. Similarly, EEG studies examining
gamma oscillatory activity in patients with schizophrenia while performing cognitive tasks have
also produced inconsistent findings. For example, reduced induced oscillatory activity has also
been reported in patients with schizophrenia while performing a cognitive control (Cho,
Konecky et al. 2006; Lewis, Cho et al. 2008) and working memory (Lewis, Cho et al. 2008;
Haenschel, Bittner et al. 2009) task. By contrast, increased evoked gamma oscillatory activity
has been reported in patients with schizophrenia (Basar-Eroglu, Brand et al. 2007) and is
consistent with the findings of studies one and three.
Divergent reduced and elevated gamma oscillatory activity findings may be accounted for
performance differences in patients with schizophrenia compared to healthy subjects. That is, in
fMRI, the parcelation of low versus high performing patients with schizophrenia during the N-
back task has revealed differences in DLPFC activation and demonstrated reduced activity in
low performers and enhanced activity in high performers (Callicott, Mattay et al. 2003). In
addition, when high performing patients with schizophrenia are compared to high performing
healthy subjects, increased activation of the DLPFC was revealed in patients (Callicott, Mattay et
al. 2003). Consistent with the findings of Callicott et al (2003), study one also compared gamma
138
oscillatory activity at equivalent performance levels and found that patients with schizophrenia
generated excessive gamma activity compared to healthy subjects. It is therefore possible that
differences due to performance in patients with schizophrenia may account for the inconsistent
fMRI and EEG findings. Furthermore, differences in EEG analytical methods may also
contribute to these divergent findings at least during cognitive tasks. That is, the studies that
employed induced EEG methods report reduced gamma activity (Cho, Konecky et al. 2006;
Lewis, Cho et al. 2008; Haenschel, Bittner et al. 2009), while excessive activity is reported with
evoked EEG methods (Basar-Eroglu, Brand et al. 2007; Barr, Farzan et al. 2010). As previously
described, these methods differ in their phase relationship with the stimulus onset. That is,
induced activity is not phase-locked to stimulus onset and appears as a jitter in latency that varies
from trial to trial and therefore possesses a loose temporal relationship with the stimulus. Evoked
oscillatory activity, by contrast, is phase-locked to the stimulus onset and is characterized by a
constant time and phase relationship with the stimulus (Tallon-Baudry and Bertrand 1999).
However, induced EEG methods have been criticized for its susceptibility to both microsaccadic
eye movement (Yuval-Greenberg, Tomer et al. 2008) and cranial musculature (Shackman 2010)
artifact. It is therefore possible that the finding of reduced gamma activity through induced EEG
methods may reflect microsaccadic or cranial musculature artifact that is canceled out when
evoked EEG methods are employed (Barr and Daskalakis 2010). The nature of prefrontal
abnormalities in relation to cognitive deficits in patients with schizophrenia continues to be
debated. Moreover, factors such as performance levels, experimental paradigms and analytical
methods, antipsychotic medication, and heterogeneity in DLPFC activation warrants further
examination in assessing gamma oscillatory activity during cognition in patients with
schizophrenia.
139
6.7 Specificity of rTMS on Frontal Evoked Gamma Oscillatory Activity in the DLPFC
In the series of studies presented in this PhD thesis we examined frontal evoked gamma
oscillatory activity in the DLPFC. We examined the DLPFC because of its role in the mediation
of working memory and the considerable evidence for this brain region in the pathophysiology of
schizophrenia (Elvevag and Goldberg 2000; Deiber, Missonnier et al. 2007). In the series of
studies presented here, rTMS was targeted at the DLPFC determined through a novel
neuronavigational technique which mapped the coordinate position of the DLPFC based on
meta-analyses on fMRI studies on working memory onto T1-weighted image of a MRI obtained
for each subject (see Appendix A; (Rusjan, Barr et al. 2010). The application of rTMS over the
DLPFC in studies two and three of this thesis resulted in the modulation of evoked oscillatory
activity in the frontal region that was absent in posterior brain regions. However, a study that
evaluated EEG coherence reported an increase in alpha power in both the frontal and the
posterior parietal cortices following rTMS over the frontal region of healthy subjects (Jing and
Takigawa 2000). Different brain regions have been shown to oscillate at their natural frequency
range and the frontal cortex has been shown to naturally oscillate at higher frequencies such as
beta and gamma compared to other brain regions (Rosanova, Casali et al. 2009). This study by
Rosanova et al (2009) may therefore explain the specificity of rTMS on frontal evoked gamma
oscillatory activity in healthy subjects. In patients with schizophrenia, though, study three
showed that rTMS reduced frontal evoked delta oscillatory activity in the 3-back condition as
well. As described earlier, interactions exist between the activities in the frequency bands due to
their non-random relationships (Roopun, Kramer et al. 2008). That is, cross-frequency or
―nesting‖ relationships describes when the power of a discrete frequency band is modified by the
phase of a lower frequency band that coexists during information processing (Roopun, Kramer et
140
al. 2008). Such nesting interactions have been observed in the hippocampus (Lakatos, Shah et al.
2005) and neocortex (Penttonen and Buzsaki 2003) between delta, theta, and gamma frequency
bands. Thus the finding of reduced frontal evoked delta and reduced gamma activity following
rTMS in patient with schizophrenia may reflect the interaction between these two frequency
bands.
6.8 Limitations and Future Directions
To our knowledge, the series of studies presented in this PhD thesis are the first to administer
rTMS to measure its effects on frontal gamma oscillatory activity generated during a subsequent
working memory task in patients with schizophrenia and healthy subjects. Although it was
demonstrated that rTMS modulated this frontal evoked gamma oscillatory activity in both
subject groups, these studies are not without their limitations. In this regard, the next section will
discuss the collective limitations to this work as specific limitations have already been discussed
within each study. In addition, future directions based on this body of work will also be explored.
First, studies two and three limited the examination of rTMS to the DLPFC, applied at 20 Hz
frequency to both hemispheres on oscillatory activity during working memory. It is possible
though that rTMS applied at a different frequency to the DLPFC may also result in the
modulation of gamma activity that may also influence working memory performance. For
example, Okamura et al demonstrated significant increases in the absolute gamma power
following 10 Hz rTMS applied over the left prefrontal cortex (Okamura, Jing et al. 2001). It has
also been reported that 10 Hz rTMS applied to the right DLPFC resulted in an increase in
accuracy on a spatial working memory task in healthy subjects (Hamidi, Tononi et al. 2009).
These studies therefore suggest that 10 Hz rTMS applied to the DLPFC results in both the
potentiation of gamma power and an improvement in working memory performance. It should
141
also be noted that these two studies administered 10 Hz rTMS to different hemispheres. In this
regard, determining the optimal rTMS site (i.e., right, left or bilateral) for the treatment of
working memory deficits should be further investigated. Hamidi et al (2009) demonstrated right
rTMS stimulation improved performance on a spatial working memory task which is consistent
with the view that the right DLPFC is recruited for spatial tasks, while the left mediates verbal
working memory (Smith and Jonides 1999). As such, it is possible that stimulation of the left
DLPFC may improve performance on verbal working memory tasks such as the N-back used in
the studies presented here. Future experiments examining optimal rTMS stimulation parameters
in addition to the DLPFC site for different working memory tasks are greatly needed.
Studies two and three also limited rTMS administration to the DLPFC and did not examine its
effect on oscillatory activity in other brain regions. The DLPFC was chosen as the optimal site in
which to stimulate based on the considerable evidence for its involvement in the mediation of
working memory, evidence for abnormal DLPFC activation during working memory tasks, and
also for the implication of this brain region in the pathophysiology of schizophrenia.
Nevertheless, neuroimaging studies have demonstrated that working memory function is not
limited to the prefrontal cortex and recruits a network involving the prefrontal cortex and the
posterior parietal cortex (Honey, Bullmore et al. 2002). It has further been suggested that
working memory information is stored in more posterior parietal regions and this information is
retrieved by the prefrontal cortex (Curtis and D'Esposito 2003; Postle 2006). These studies
therefore suggest that rTMS applied to the posterior parietal cortex may also improve working
memory function. In this regard, 10 Hz rTMS applied to the superior parietal lobule (Hamidi,
Tononi et al. 2008) and 5 Hz rTMS applied to the parietal cortex (Luber, Kinnunen et al. 2007)
has been shown to result in a significant reduction in reaction time in healthy subjects tested in a
working memory task. Taken together, future studies that examine optimal sites for rTMS
142
stimulation in addition to the effect of different stimulation parameters are needed in the
continual investigation of rTMS as a potential treatment option for working memory deficits in
patients with schizophrenia.
Another limitation of studies two and three presented here relates to the evaluation of oscillatory
activity following rTMS administration. Although it was demonstrated that rTMS modulated
gamma oscillatory activity during a subsequent testing of the N-back administered within 20
minutes of the stimulation, it would be of interest to examine the effect of rTMS during working
memory performance. In this regard, Hamidi et al (2010) recently examined the effect of 10 Hz
stimulation applied to the postcentral gyrus and superior parietal lobule in healthy subjects while
performing a working memory task (Hamidi, Slagter et al. 2010). Through independent
component analysis, Hamidi argued that rTMS related artifact can be removed without
influencing neurophysiological activity as electrical artifact is temporally and spatially
predictable (Hamidi, Slagter et al. 2010). In this study, a quadratic relationship was demonstrated
between potential peak amplitude and pulse number within the TMS train which was
characterized by a decrease followed by an increase in amplitude. That is, rTMS was shown to
result in both the depression and potentiation of neural excitability and thus the effects of rTMS
are more complex than the simple linear summation of the neural response (Hamidi, Slagter et al.
2010). Hamidi et al are the first to deliver rTMS while recording EEG activity and demonstrate
through independent component analysis, rTMS related artifacts can be removed. This study
underscores the need to evaluate pulse to pulse changes induced by rTMS and whether these
neurophysiological changes are related to stimulation frequency. Such examination of rTMS
effects on-line will thus provide a greater understanding of the effects induced by rTMS on
oscillatory activity in the subjects tested in studies two and three.
143
A final limitation of these studies relates to the paradigm used to assess working memory. We
administered the N-back task which is a challenging working memory task however it does not
allow for the examination of subprocesses of working memory. That is, despite the numerous
theories that exist on working memory, it is accepted that this mechanism includes encoding,
maintenance, manipulation, and retrieval subprocesses. In this regard, previous studies have
shown alterations in oscillatory activity within these subprocesses (e.g., (Haenschel, Bittner et al.
2009; Ince, Pellizzer et al. 2009)). Future studies which evaluate the effect of rTMS on
oscillatory activity specifically generated within working memory subprocesses may provide a
better understanding of how to optimize rTMS as a potential tool in which to ameliorate working
memory deficits in patients with schizophrenia.
6.9 Conclusion
The series of studies presented in this PhD thesis demonstrated that patients with schizophrenia
generate excessive frontal evoked gamma oscillatory activity during the N-back task that may
contribute to working memory impairment in this disorder. Furthermore, it was demonstrated
that high frequency rTMS over the DLPFC inhibits excessive frontal evoked gamma oscillatory
activity in patients with schizophrenia while potentiating this activity in healthy subjects when
tested in a subsequent N-back test. Although these studies are not without their limitations, they
provide a better understanding of the neurophysiological mechanisms that may contribute to
working memory deficits in patients with schizophrenia and suggest that rTMS may be used as a
potential cognitive enhancing tool in which to ameliorate these deficits possibly through repeated
treatment sessions.
144
References
Adcock, R. A., C. Dale, et al. (2009). "When top-down meets bottom-up: auditory training
enhances verbal memory in schizophrenia." Schizophr Bull 35(6): 1132-41.
Addington, D., J. Addington, et al. (1993). "Assessing depression in schizophrenia: the Calgary
Depression Scale." Br J Psychiatry Suppl(22): 39-44.
Akbarian, S., J. J. Kim, et al. (1995). "Gene expression for glutamic acid decarboxylase is
reduced without loss of neurons in prefrontal cortex of schizophrenics [see comments]."
Arch Gen Psychiatry 52(4): 258-66; discussion 267-78.
Amati, F., E. Conti, et al. (1999). "Atypical deletions suggest five 22q11.2 critical regions related
to the DiGeorge/velo-cardio-facial syndrome." Eur J Hum Genet 7(8): 903-9.
Andlin-Sobocki, P. and W. Rossler (2005). "Cost of psychotic disorders in Europe." Eur J
Neurol 12 Suppl 1: 74-7.
Andoh, J., D. Riviere, et al. (2009). "A triangulation-based magnetic resonance image-guided
method for transcranial magnetic stimulation coil positioning." Brain Stimulation In
Press.
Andreasen, N. C. (1989). "The Scale for the Assessment of Negative Symptoms (SANS):
conceptual and theoretical foundations." Br J Psychiatry Suppl(7): 49-58.
Angrist, B. M. and D. P. van Kammen (1984). "CNS stimulants as a tool in the study of
schizophrenia." Trends Neurosci 7: 388-390.
Appels, M. C., M. M. Sitskoorn, et al. (2003). "Cognitive dysfunctions in parents of
schizophrenic patients parallel the deficits found in patients." Schizophr Res 63(3): 285-
93.
Arnsten, A. F., J. X. Cai, et al. (1995). "Dopamine D2 receptor mechanisms contribute to age-
related cognitive decline: the effects of quinpirole on memory and motor performance in
monkeys." J Neurosci 15(5 Pt 1): 3429-39.
Arnsten, A. F. and T. A. Contant (1992). "Alpha-2 adrenergic agonists decrease distractibility in
aged monkeys performing the delayed response task." Psychopharmacology (Berl)
108(1-2): 159-69.
Arnsten, A. F. and P. S. Goldman-Rakic (1998). "Noise stress impairs prefrontal cortical
cognitive function in monkeys: evidence for a hyperdopaminergic mechanism." Arch
Gen Psychiatry 55(4): 362-8.
Ashburner, J. and K. J. Friston (1999). "Nonlinear spatial normalization using basis functions."
Hum Brain Mapp 7(4): 254-66.
Ashburner, J., P. Neelin, et al. (1997). "Incorporating prior knowledge into image registration."
Neuroimage 6(4): 344-52.
145
Bachman, P., J. Kim, et al. (2008). "Abnormally high EEG alpha synchrony during working
memory maintenance in twins discordant for schizophrenia." Schizophr Res 103(1-3):
293-7.
Baddeley, A. (1986). Working memory. Oxford, Claredon Press.
Baddeley, A. (1992). "Working memory." Science 255(5044): 556-9.
Baddeley, A. (2000). "The episodic buffer: a new component of working memory?" Trends
Cogn Sci 4(11): 417-423.
Baddeley, A. D. and G. Hitch (1974). Working Memory. The psychology of learning and
motivation: Advances in research and theory. G. H. Bower. New York, Academic Press.
8: 47-89.
Bagnato, S., A. Curra, et al. (2005). "One-hertz subthreshold rTMS increases the threshold for
evoking inhibition in the human motor cortex." Exp Brain Res 160(3): 368-74.
Barbas, H. (1988). "Anatomic organization of basoventral and mediodorsal visual recipient
prefrontal regions in the rhesus monkey." J Comp Neurol 276(3): 313-42.
Barch, D. M. and C. S. Carter (2005). "Amphetamine improves cognitive function in medicated
individuals with schizophrenia and in healthy volunteers." Schizophr Res 77(1): 43-58.
Barch, D. M., C. S. Carter, et al. (2001). "Selective deficits in prefrontal cortex function in
medication-naive patients with schizophrenia." Arch Gen Psychiatry 58(3): 280-8.
Barch, D. M., J. G. Csernansky, et al. (2002). "Working and long-term memory deficits in
schizophrenia: is there a common prefrontal mechanism?" J Abnorm Psychol 111(3):
478-94.
Barch, D. M. and E. Smith (2008). "The cognitive neuroscience of working memory: relevance
to CNTRICS and schizophrenia." Biol Psychiatry 64(1): 11-7.
Barden, N. and J. Morissette (1999). "Chromosome 13 workshop report." Am J Med Genet
88(3): 260-2.
Barr, M. S. and Z. J. Daskalakis (2010). "In response to: the potentially deleterious impact of
muscle activity on gamma band inferences." Neuropsychopharmacology 35: 848-849.
Barr, M. S., F. Farzan, et al. (2009). "Potentiation of gamma oscillatory activity through
repetitive transcranial magnetic stimulation of the dorsolateral prefrontal cortex."
Neuropsychopharmacology 34(11): 2359-67.
Barr, M. S., F. Farzan, et al. (2010). "Evidence for excessive frontal evoked gamma oscillatory
activity in schizophrenia during working memory." Schizophr Res 121(1-3): 146-52.
146
Barr, M. S., Fitzgerald, P.B., Farzan, F, George, T.P., Daskalakis, Z.J. (2008). "Transcranial
magnetic stimulation to understand the pathophysiology and treatment of substance use
disorders." Current Drug Abuse Reviews 1(3): 328-339.
Bartos, M., I. Vida, et al. (2007). "Synaptic mechanisms of synchronized gamma oscillations in
inhibitory interneuron networks." Nat Rev Neurosci 8(1): 45-56.
Basar-Eroglu, C., A. Brand, et al. (2007). "Working memory related gamma oscillations in
schizophrenia patients." Int J Psychophysiol 64(1): 39-45.
Bassett, A. S. (1992). "Chromosomal aberrations and schizophrenia. Autosomes." Br J
Psychiatry 161: 323-34.
Bassett, A. S., E. W. Chow, et al. (2000). "Chromosomal abnormalities and schizophrenia." Am
J Med Genet 97(1): 45-51.
Basso, M. R., H. A. Nasrallah, et al. (1998). "Neuropsychological correlates of negative,
disorganized and psychotic symptoms in schizophrenia." Schizophr Res 31(2-3): 99-111.
Bellivier, F., A. Szoke, et al. (2000). "Possible association between serotonin transporter gene
polymorphism and violent suicidal behavior in mood disorders." Biol Psychiatry 48(4):
319-22.
Ben-Shachar, D., R. H. Belmaker, et al. (1997). "Transcranial magnetic stimulation induces
alterations in brain monoamines." J Neural Transm 104(2-3): 191-7.
Bender, R., B. Schultz, et al. (1992). "Testing the Gaussianity of the human EEG during
anesthesia." Methods Inf Med 31(1): 56-9.
Benes, F. M. (1997). "The role of stress and dopamine-GABA interactions in the vulnerability
for schizophrenia." J Psychiatr Res 31(2): 257-75.
Benes, F. M. (1999). "Evidence for altered trisynaptic circuitry in schizophrenic hippocampus."
Biol Psychiatry 46(5): 589-99.
Benes, F. M. and S. Berretta (2001). "GABAergic interneurons: implications for understanding
schizophrenia and bipolar disorder." Neuropsychopharmacology 25(1): 1-27.
Benes, F. M., J. McSparren, et al. (1991). "Deficits in small interneurons in prefrontal and
cingulate cortices of schizophrenic and schizoaffective patients." Arch Gen Psychiatry
48(11): 996-1001.
Bertolino, A. and G. Blasi (2009). "The genetics of schizophrenia." Neuroscience 164(1): 288-
99.
Bilder, R. M., R. S. Goldman, et al. (2002). "Neurocognitive effects of clozapine, olanzapine,
risperidone, and haloperidol in patients with chronic schizophrenia or schizoaffective
disorder." Am J Psychiatry 159(6): 1018-28.
147
Blanchard, J. J. and J. M. Neale (1994). "The neuropsychological signature of schizophrenia:
generalized or differential deficit?" Am J Psychiatry 151(1): 40-8.
Boggio, P. S., F. Fregni, et al. (2005). "Effect of repetitive TMS and fluoxetine on cognitive
function in patients with Parkinson's disease and concurrent depression." Mov Disord
20(9): 1178-84.
Bora, E., B. Veznedaroglu, et al. (2005). "The effect of galantamine added to clozapine on
cognition of five patients with schizophrenia." Clin Neuropharmacol 28(3): 139-41.
Botstein, D., R. L. White, et al. (1980). "Construction of a genetic linkage map in man using
restriction fragment length polymorphisms." Am J Hum Genet 32(3): 314-31.
Braver, T. S., D. M. Barch, et al. (1999). "Cognition and control in schizophrenia: a
computational model of dopamine and prefrontal function." Biol Psychiatry 46(3): 312-
28.
Braver, T. S. and J. D. Cohen (1999). "Dopamine, cognitive control, and schizophrenia: the
gating model." Prog Brain Res 121: 327-49.
Braver, T. S., J. D. Cohen, et al. (1997). "A parametric study of prefrontal cortex involvement in
human working memory." Neuroimage 5(1): 49-62.
Brookes, M. J., A. M. Gibson, et al. (2005). "GLM-beamformer method demonstrates stationary
field, alpha ERD and gamma ERS co-localisation with fMRI BOLD response in visual
cortex." Neuroimage 26(1): 302-8.
Brovelli, A., J. P. Lachaux, et al. (2005). "High gamma frequency oscillatory activity dissociates
attention from intention in the human premotor cortex." Neuroimage 28(1): 154-64.
Brown, J. T., C. H. Davies, et al. (2007). "Synaptic activation of GABA(B) receptors regulates
neuronal network activity and entrainment." Eur J Neurosci 25(10): 2982-90.
Brzustowicz, L. M., K. A. Hodgkinson, et al. (2000). "Location of a major susceptibility locus
for familial schizophrenia on chromosome 1q21-q22." Science 288(5466): 678-82.
Burt, T., S. H. Lisanby, et al. (2002). "Neuropsychiatric applications of transcranial magnetic
stimulation: a meta analysis." Int J Neuropsychopharmacol 5(1): 73-103.
Buzsaki, G. (2006). Rhythms of the Brain. New York, New York, Oxford University Press.
Buzsaki, G. and A. Draguhn (2004). "Neuronal oscillations in cortical networks." Science
304(5679): 1926-9.
Callicott, J. H., A. Bertolino, et al. (2000). "Physiological dysfunction of the dorsolateral
prefrontal cortex in schizophrenia revisited." Cereb Cortex 10(11): 1078-92.
Callicott, J. H., V. S. Mattay, et al. (1999). "Physiological characteristics of capacity constraints
in working memory as revealed by functional MRI." Cereb Cortex 9(1): 20-6.
148
Callicott, J. H., V. S. Mattay, et al. (2003). "Complexity of prefrontal cortical dysfunction in
schizophrenia: more than up or down." Am J Psychiatry 160(12): 2209-15.
Callicott, J. H., N. F. Ramsey, et al. (1998). "Functional magnetic resonance imaging brain
mapping in psychiatry: methodological issues illustrated in a study of working memory in
schizophrenia." Neuropsychopharmacology 18(3): 186-96.
Cannon, M., P. B. Jones, et al. (2002). "Obstetric complications and schizophrenia: historical and
meta-analytic review." Am J Psychiatry 159(7): 1080-92.
Cannon, T. D., D. C. Glahn, et al. (2005). "Dorsolateral prefrontal cortex activity during
maintenance and manipulation of information in working memory in patients with
schizophrenia." Arch Gen Psychiatry 62(10): 1071-80.
Cannon, T. D., L. E. Zorrilla, et al. (1994). "Neuropsychological functioning in siblings
discordant for schizophrenia and healthy volunteers." Arch Gen Psychiatry 51(8): 651-
61.
Carlsson, A. and M. Lindqvist (1963). "Effect of Chlorpromazine or Haloperidol on Formation
of 3methoxytyramine and Normetanephrine in Mouse Brain." Acta Pharmacol Toxicol
(Copenh) 20: 140-4.
Castle, D. J., S. Wessely, et al. (1993). "Sex and schizophrenia: effects of diagnostic stringency,
and associations with and premorbid variables." Br J Psychiatry 162: 658-64.
Cavada, C. and P. S. Goldman-Rakic (1989). "Posterior parietal cortex in rhesus monkey: II.
Evidence for segregated corticocortical networks linking sensory and limbic areas with
the frontal lobe." J Comp Neurol 287(4): 422-45.
Celada, P., M. V. Puig, et al. (2008). "The hallucinogen DOI reduces low-frequency oscillations
in rat prefrontal cortex: reversal by antipsychotic drugs." Biol Psychiatry 64(5): 392-400.
Chang, S. M., S. J. Cho, et al. (2008). "Economic burden of schizophrenia in South Korea." J
Korean Med Sci 23(2): 167-75.
Chapman, L. J. and J. P. Chapman (1973). "Problems in the measurement of cognitive deficit."
Psychol Bull 79(6): 380-5.
Chen, N., X. Chen, et al. (2008). "Homeostasis established by coordination of subcellular
compartment plasticity improves spike encoding." J Cell Sci 121(Pt 17): 2961-71.
Chen, R., C. Gerloff, et al. (1997). "Safety of different inter-train intervals for repetitive
transcranial magnetic stimulation and recommendations for safe ranges of stimulation
parameters." Electroencephalogr Clin Neurophysiol 105(6): 415-21.
Chen, X., S. Shu, et al. (2010). "Homeostatic regulation of synaptic excitability: tonic GABA(A)
receptor currents replace I(h) in cortical pyramidal neurons of HCN1 knock-out mice." J
Neurosci 30(7): 2611-22.
149
Chisholm, D., O. Gureje, et al. (2008). "Schizophrenia treatment in the developing world: an
interregional and multinational cost-effectiveness analysis." Bull World Health Organ
86(7): 542-51.
Cho, R. Y., R. O. Konecky, et al. (2006). "Impairments in frontal cortical gamma synchrony and
cognitive control in schizophrenia." Proc Natl Acad Sci U S A 103(52): 19878-83.
Cobb, S. R., E. H. Buhl, et al. (1995). "Synchronization of neuronal activity in hippocampus by
individual GABAergic interneurons." Nature 378(6552): 75-8.
Cohen, J. D., W. M. Perlstein, et al. (1997). "Temporal dynamics of brain activation during a
working memory task." Nature 386(6625): 604-8.
Collin, T., M. Chat, et al. (2005). "Developmental changes in parvalbumin regulate presynaptic
Ca2+ signaling." J Neurosci 25(1): 96-107.
Conca, A., S. Koppi, et al. (1996). "Transcranial magnetic stimulation: a novel antidepressive
strategy?" Neuropsychobiology 34(4): 204-7.
Constantinidis, C., G. V. Williams, et al. (2002). "A role for inhibition in shaping the temporal
flow of information in prefrontal cortex." Nat Neurosci 5(2): 175-80.
Coon, H., S. Jensen, et al. (1994). "Genomic scan for genes predisposing to schizophrenia." Am
J Med Genet 54(1): 59-71.
Cosway, R., M. Byrne, et al. (2000). "Neuropsychological change in young people at high risk
for schizophrenia: results from the first two neuropsychological assessments of the
Edinburgh High Risk Study." Psychol Med 30(5): 1111-21.
Coull, J. T., H. C. Middleton, et al. (1995). "Contrasting effects of clonidine and diazepam on
tests of working memory and planning." Psychopharmacology (Berl) 120(3): 311-21.
Courtney, S. M., L. Petit, et al. (1998). "An area specialized for spatial working memory in
human frontal cortex." Science 279(5355): 1347-51.
Courtney, S. M., L. G. Ungerleider, et al. (1997). "Transient and sustained activity in a
distributed neural system for human working memory." Nature 386(6625): 608-11.
Cowan, N. (1997). Attention and Memory: An integrated Framework. New York, Oxford
University Press.
Cowan, N. (2001). "The magical number 4 in short-term memory: a reconsideration of mental
storage capacity." Behav Brain Sci 24: 87-114.
Craddock, N. and C. Lendon (1999). "Chromosome Workshop: chromosomes 11, 14, and 15."
Am J Med Genet 88(3): 244-54.
Creese, I., D. R. Burt, et al. (1976). "Dopamine receptor binding predicts clinical and
pharmacological potencies of antischizophrenic drugs." Science 192(4238): 481-3.
150
Crowe, R. R. and V. Vieland (1999). "Report of the Chromosome 5 Workshop of the Sixth
World Congress on Psychiatric Genetics." Am J Med Genet 88(3): 229-32.
Curtis, C. E., M. E. Calkins, et al. (2001). "Saccadic disinhibition in patients with acute and
remitted schizophrenia and their first-degree biological relatives." Am J Psychiatry
158(1): 100-6.
Curtis, C. E. and M. D'Esposito (2003). "Persistent activity in the prefrontal cortex during
working memory." Trends Cogn Sci 7(9): 415-423.
Curtis, C. E. and M. D'Esposito (2004). "The effects of prefrontal lesions on working memory
performance and theory." Cogn Affect Behav Neurosci 4(4): 528-39.
Curtis, D. (1999). "Chromosome 21 workshop." Am J Med Genet 88(3): 272-5.
D'Esposito, M. (2007). "From cognitive to neural models of working memory." Philos Trans R
Soc Lond B Biol Sci 362(1481): 761-72.
D'Esposito, M., G. K. Aguirre, et al. (1998). "Functional MRI studies of spatial and nonspatial
working memory." Brain Res Cogn Brain Res 7(1): 1-13.
Daneman, M. and T. Tardif (1987). Working memory and reading skill reexamined. Attention
and performance XII: The psychology of reading. M. Coltheart. Hillsdale, N.J., Erlbaum.
Daniel, D. G., K. F. Berman, et al. (1989). "The effect of apomorphine on regional cerebral
blood flow in schizophrenia." J Neuropsychiatry Clin Neurosci 1(4): 377-84.
Daskalakis, Z. J., B. K. Christensen, et al. (2002). "Evidence for impaired cortical inhibition in
schizophrenia using transcranial magnetic stimulation." Arch Gen Psychiatry 59(4): 347-
54.
Daskalakis, Z. J., B. K. Christensen, et al. (2008). "Dysfunctional neural plasticity in patients
with schizophrenia." Arch Gen Psychiatry 65(4): 378-85.
Daskalakis, Z. J., B. K. Christensen, et al. (2008). "Increased cortical inhibition in persons with
schizophrenia treated with clozapine." J Psychopharmacol 22(2): 203-9.
Daskalakis, Z. J., F. Farzan, et al. (2008). "Long-interval cortical inhibition from the dorsolateral
prefrontal cortex: a TMS-EEG study." Neuropsychopharmacology 33(12): 2860-9.
Daskalakis, Z. J., B. Moller, et al. (2006). "The effects of repetitive transcranial magnetic
stimulation on cortical inhibition in healthy human subjects." Exp Brain Res 174(3): 403-
12.
David, O., J. M. Kilner, et al. (2006). "Mechanisms of evoked and induced responses in
MEG/EEG." Neuroimage 31(4): 1580-91.
Davis, K. L., R. S. Kahn, et al. (1991). "Dopamine in schizophrenia: a review and
reconceptualization." Am J Psychiatry 148(11): 1474-86.
151
DeFelipe, J. and I. Farinas (1992). "The pyramidal neuron of the cerebral cortex: morphological
and chemical characteristics of the synaptic inputs." Prog Neurobiol 39(6): 563-607.
Deiber, M. P., P. Missonnier, et al. (2007). "Distinction between perceptual and attentional
processing in working memory tasks: a study of phase-locked and induced oscillatory
brain dynamics." J Cogn Neurosci 19(1): 158-72.
Delorme, A. and S. Makeig (2004). "EEGLAB: an open source toolbox for analysis of single-
trial EEG dynamics including independent component analysis." J Neurosci Methods
134(1): 9-21.
Detera-Wadleigh, S. D. (1999). "Chromosomes 12 and 16 workshop." Am J Med Genet 88(3):
255-9.
Dickinson, D., W. Tenhula, et al. (2010). "A randomized, controlled trial of computer-assisted
cognitive remediation for schizophrenia." Am J Psychiatry 167(2): 170-80.
Druzgal, T. J. and M. D'Esposito (2003). "Dissecting contributions of prefrontal cortex and
fusiform face area to face working memory." J Cogn Neurosci 15(6): 771-84.
Duncan, G. E., S. Miyamoto, et al. (1999). "Comparison of brain metabolic activity patterns
induced by ketamine, MK-801 and amphetamine in rats: support for NMDA receptor
involvement in responses to subanesthetic dose of ketamine." Brain Res 843(1-2): 171-
83.
Durstewitz, D., J. K. Seamans, et al. (2000). "Dopamine-mediated stabilization of delay-period
activity in a network model of prefrontal cortex." J Neurophysiol 83(3): 1733-50.
Egan, M. F., T. E. Goldberg, et al. (2001). "Effect of COMT Val108/158 Met genotype on
frontal lobe function and risk for schizophrenia." Proc Natl Acad Sci U S A 98(12):
6917-22.
Eisenegger, C., V. Treyer, et al. (2008). "Time-course of "off-line" prefrontal rTMS effects--a
PET study." Neuroimage 42(1): 379-84.
Ekelund, J., I. Hovatta, et al. (2001). "Chromosome 1 loci in Finnish schizophrenia families."
Hum Mol Genet 10(15): 1611-7.
Elvevag, B. and T. E. Goldberg (2000). "Cognitive impairment in schizophrenia is the core of
the disorder." Crit Rev Neurobiol 14(1): 1-21.
Engle, R. W., J. Cantor, et al. (1992). "Individual differences in working memory and
comprehension: a test of four hypotheses." J Exp Psychol Learn Mem Cogn 18(5): 972-
92.
Erhardt, A., I. Sillaber, et al. (2004). "Repetitive transcranial magnetic stimulation increases the
release of dopamine in the nucleus accumbens shell of morphine-sensitized rats during
abstinence." Neuropsychopharmacology 29(11): 2074-80.
152
Ericsson, K. A. and W. Kintsch (1995). "Long-term working memory." Psychol Rev 102(2):
211-45.
Escher, T. and G. Mittleman (2004). "Effects of ethanol and GABAB drugs on working memory
in C57BL/6J and DBA/2J mice." Psychopharmacology (Berl) 176(2): 166-74.
Evans, A. C., Collins, D.L. (1993). A 305-member MRI-based sterotactic atlas for CBF
activation studies. Proceedings of the 40th annual meeting of the Society for Nuclear
Medicine.
Faraone, S. V., L. J. Seidman, et al. (1996). "Neuropsychological functioning among the elderly
nonpsychotic relatives of schizophrenic patients." Schizophr Res 21(1): 27-31.
Farzan, F., M. S. Barr, et al. (2010). "Evidence for gamma inhibition deficits in the dorsolateral
prefrontal cortex of patients with schizophrenia." Brain 133(Pt 5): 1505-14.
Fioravanti, M., O. Carlone, et al. (2005). "A meta-analysis of cognitive deficits in adults with a
diagnosis of schizophrenia." Neuropsychol Rev 15(2): 73-95.
Fisahn, A., F. G. Pike, et al. (1998). "Cholinergic induction of network oscillations at 40 Hz in
the hippocampus in vitro." Nature 394(6689): 186-9.
Fisher, M., C. Holland, et al. (2009). "Using neuroplasticity-based auditory training to improve
verbal memory in schizophrenia." Am J Psychiatry 166(7): 805-11.
Fitzgerald, P. B., J. Benitez, et al. (2005). "A double-blind sham-controlled trial of repetitive
transcranial magnetic stimulation in the treatment of refractory auditory hallucinations." J
Clin Psychopharmacol 25(4): 358-62.
Fitzgerald, P. B., T. L. Brown, et al. (2002). "A transcranial magnetic stimulation study of the
effects of olanzapine and risperidone on motor cortical excitability in patients with
schizophrenia." Psychopharmacology (Berl) 162(1): 74-81.
Fitzgerald, P. B., K. Hoy, et al. (2009). "A randomized trial of rTMS targeted with MRI based
neuro-navigation in treatment-resistant depression." Neuropsychopharmacology 34(5):
1255-62.
Fitzgerald, P. B., J. J. Maller, et al. (2009). "Exploring the optimal site for the localization of
dorsolateral prefrontal cortex in brain stimulation experiments." Brain Stimulation In
Press.
Fitzgerald, P. B., T. J. Oxley, et al. (2006). "An analysis of functional neuroimaging studies of
dorsolateral prefrontal cortical activity in depression." Psychiatry Res 148(1): 33-45.
Fleischmann, A., K. Prolov, et al. (1995). "The effect of transcranial magnetic stimulation of rat
brain on behavioral models of depression." Brain Res 699(1): 130-2.
Fletcher, P. C. and R. N. Henson (2001). "Frontal lobes and human memory: insights from
functional neuroimaging." Brain 124(Pt 5): 849-81.
153
Forbes, N. F., L. A. Carrick, et al. (2009). "Working memory in schizophrenia: a meta-analysis."
Psychol Med 39(6): 889-905.
Foucher, J. R., H. Otzenberger, et al. (2003). "The BOLD response and the gamma oscillations
respond differently than evoked potentials: an interleaved EEG-fMRI study." BMC
Neurosci 4: 22.
Friedman, H. R. and P. S. Goldman-Rakic (1994). "Coactivation of prefrontal cortex and inferior
parietal cortex in working memory tasks revealed by 2DG functional mapping in the
rhesus monkey." J Neurosci 14(5 Pt 1): 2775-88.
Fries, P., D. Nikolic, et al. (2007). "The gamma cycle." Trends Neurosci 30(7): 309-16.
Fries, P., J. H. Reynolds, et al. (2001). "Modulation of oscillatory neuronal synchronization by
selective visual attention." Science 291(5508): 1560-3.
Frisk, V. and B. Milner (1990). "The relationship of working memory to the immediate recall of
stories following unilateral temporal or frontal lobectomy." Neuropsychologia 28(2):
121-35.
Furey, M. L., P. Pietrini, et al. (2000). "Cholinergic enhancement improves performance on
working memory by modulating the functional activity in distinct brain regions: a
positron emission tomography regional cerebral blood flow study in healthy humans."
Brain Res Bull 51(3): 213-8.
Furey, M. L., P. Pietrini, et al. (2000). "Cholinergic enhancement and increased selectivity of
perceptual processing during working memory." Science 290(5500): 2315-9.
Fuster, J. M. (1995). "Memory in the cortex of the primate." Biol Res 28(1): 59-72.
Fuster, J. M., R. H. Bauer, et al. (1985). "Functional interactions between inferotemporal and
prefrontal cortex in a cognitive task." Brain Res 330(2): 299-307.
Gaddum, J. H. (1954). Drug antagonistic to 5-hydroxytryptamine. . Ciba Fundation Symposium
on Hypertension. G. W. Wolstenholme. Boston, Little Brown and Company: 75–77.
Galarreta, M. and S. Hestrin (1999). "A network of fast-spiking cells in the neocortex connected
by electrical synapses." Nature 402(6757): 72-5.
Gejman, P. V. (1999). "Chromosomes 19 and 20 report." Am J Med Genet 88(3): 271.
George, M. S., E. M. Wassermann, et al. (1995). "Daily repetitive transcranial magnetic
stimulation (rTMS) improves mood in depression." Neuroreport 6(14): 1853-6.
Gerloff, C., B. Corwell, et al. (1997). "Stimulation over the human supplementary motor area
interferes with the organization of future elements in complex motor sequences." Brain
120 ( Pt 9): 1587-602.
154
Gershon, A. A., P. N. Dannon, et al. (2003). "Transcranial magnetic stimulation in the treatment
of depression." Am J Psychiatry 160(5): 835-45.
Gevins, A., M. E. Smith, et al. (1997). "High-resolution EEG mapping of cortical activation
related to working memory: effects of task difficulty, type of processing, and practice."
Cereb Cortex 7(4): 374-85.
Gibson, J. F., M. Beierlein, et al. (1999). "Two networks of electrically coupled inhibitory
neurons in the neocortex." Nature 402: 75-79.
Glahn, D. C., J. D. Ragland, et al. (2005). "Beyond hypofrontality: a quantitative meta-analysis
of functional neuroimaging studies of working memory in schizophrenia." Hum Brain
Mapp 25(1): 60-9.
Glass, L. (2001). "Synchronization and rhythmic processes in physiology." Nature 410(6825):
277-84.
Goeree, R., F. Farahati, et al. (2005). "The economic burden of schizophrenia in Canada in
2004." Curr Med Res Opin 21(12): 2017-28.
Gold, J. M. (2004). "Cognitive deficits as treatment targets in schizophrenia." Schizophr Res
72(1): 21-8.
Goldberg, T. E. and D. R. Weinberger (2004). "Genes and the parsing of cognitive processes."
Trends Cogn Sci 8(7): 325-35.
Goldman-Rakic, P. S. (1988). "Topography of cognition: parallel distributed networks in primate
association cortex." Annu Rev Neurosci 11: 137-56.
Goldman-Rakic, P. S. (1992). Working memory and the mind. The Scientific American book of
the brain. S. A. Inc. Guilford Connecticut: 91-104.
Goldman-Rakic, P. S. (1995). "Cellular basis of working memory." Neuron 14(3): 477-85.
Goldman-Rakic, P. S., E. C. Muly, 3rd, et al. (2000). "D(1) receptors in prefrontal cells and
circuits." Brain Res Brain Res Rev 31(2-3): 295-301.
Gonzalez-Burgos, G. and D. A. Lewis (2008). "GABA neurons and the mechanisms of network
oscillations: implications for understanding cortical dysfunction in schizophrenia."
Schizophr Bull 34(5): 944-61.
Gottesman, I. I. and J. Shields (1972). Schizophrenia and Genetics: A Twin Study Vantage Point.
New York, Academic Press.
Gray, C. M., P. Konig, et al. (1989). "Oscillatory responses in cat visual cortex exhibit inter-
columnar synchronization which reflects global stimulus properties." Nature 338(6213):
334-7.
155
Green, M. F. (1996). "What are the functional consequences of neurocognitive deficits in
schizophrenia?" Am J Psychiatry 153(3): 321-30.
Griskova, I., O. Ruksenas, et al. (2007). "The effects of 10 Hz repetitive transcranial magnetic
stimulation on resting EEG power spectrum in healthy subjects." Neurosci Lett 419(2):
162-7.
Grohmann, M. and U. Trendelenburg (1988). "The handling of five amines by the extraneuronal
deaminating system of the rat heart." Naunyn Schmiedebergs Arch Pharmacol 337(2):
159-63.
Grove, W. M., B. S. Lebow, et al. (1991). "Familial prevalence and coaggregation of schizotypy
indicators: a multitrait family study." J Abnorm Psychol 100(2): 115-21.
Gruber, T. and M. M. Muller (2005). "Oscillatory brain activity dissociates between associative
stimulus content in a repetition priming task in the human EEG." Cereb Cortex 15(1):
109-16.
Gupta, A., Y. Wang, et al. (2000). "Organizing principles for a diversity of GABAergic
interneurons and synapses in the neocortex." Science 287(5451): 273-8.
Gurevich, E. V. and J. N. Joyce (1997). "Alterations in the cortical serotonergic system in
schizophrenia: a postmortem study." Biol Psychiatry 42(7): 529-45.
Haenschel, C., R. A. Bittner, et al. (2009). "Cortical oscillatory activity is critical for working
memory as revealed by deficits in early-onset schizophrenia." J Neurosci 29(30): 9481-9.
Hajak, G., J. Marienhagen, et al. (2004). "High-frequency repetitive transcranial magnetic
stimulation in schizophrenia: a combined treatment and neuroimaging study." Psychol
Med 34(7): 1157-63.
Hallmayer, J. (1999). "Chromosomes 1, 2, and 7 workshop." Am J Med Genet 88(3): 219-23.
Hallmayer, J., W. Maier, et al. (1994). "No evidence of linkage between the dopamine D2
receptor gene and schizophrenia." Psychiatry Res 53(2): 203-15.
Hamada, K. and J. A. Wada (1998). "Amygdaloid kindling in brainstem bisected cats." Epilepsy
Res 29(2): 87-95.
Hamidi, M., H. A. Slagter, et al. (2010). "Brain responses evoked by high-frequency repetitive
transcranial magnetic stimulation: an event-related potential study." Brain Stimul 3(1): 2-
14.
Hamidi, M., G. Tononi, et al. (2008). "Evaluating frontal and parietal contributions to spatial
working memory with repetitive transcranial magnetic stimulation." Brain Res 1230:
202-10.
156
Hamidi, M., G. Tononi, et al. (2009). "Evaluating the role of prefrontal and parietal cortices in
memory-guided response with repetitive transcranial magnetic stimulation."
Neuropsychologia 47(2): 295-302.
Haraldsson, H. M., F. Ferrarelli, et al. (2004). "Transcranial Magnetic Stimulation in the
investigation and treatment of schizophrenia: a review." Schizophr Res 71(1): 1-16.
Harding, C. M., G. W. Brooks, et al. (1987). "The Vermont longitudinal study of persons with
severe mental illness, II: Long-term outcome of subjects who retrospectively met DSM-
III criteria for schizophrenia." Am J Psychiatry 144(6): 727-35.
Hartmann, K., C. Bruehl, et al. (2008). "Fast homeostatic plasticity of inhibition via activity-
dependent vesicular filling." PLoS One 3(8): e2979.
Harvey, P. D., E. Howanitz, et al. (1998). "Symptoms, cognitive functioning, and adaptive skills
in geriatric patients with lifelong schizophrenia: a comparison across treatment sites."
Am J Psychiatry 155(8): 1080-6.
Harvey, P. D., E. Sacchetti, et al. (2008). "A randomized double-blind comparison of ziprasidone
vs. clozapine for cognition in patients with schizophrenia selected for resistance or
intolerance to previous treatment." Schizophr Res 105(1-3): 138-43.
Hashimoto, T., H. H. Bazmi, et al. (2008). "Conserved regional patterns of GABA-related
transcript expression in the neocortex of subjects with schizophrenia." Am J Psychiatry
165(4): 479-89.
Hashimoto, T., N. Nishino, et al. (1991). "Increase in serotonin 5-HT1A receptors in prefrontal
and temporal cortices of brains from patients with chronic schizophrenia." Life Sci 48(4):
355-63.
Hashimoto, T., D. W. Volk, et al. (2003). "Gene expression deficits in a subclass of GABA
neurons in the prefrontal cortex of subjects with schizophrenia." J Neurosci 23(15): 6315-
26.
Hazy, T. E., M. J. Frank, et al. (2007). "Towards an executive without a homunculus:
computational models of the prefrontal cortex/basal ganglia system." Philos Trans R Soc
Lond B Biol Sci 362(1485): 1601-13.
Heinrichs, R. W. (2005). "The primacy of cognition in schizophrenia." Am Psychol 60(3): 229-
42.
Heinrichs, R. W. and K. K. Zakzanis (1998). "Neurocognitive deficit in schizophrenia: a
quantitative review of the evidence." Neuropsychology 12(3): 426-45.
Henseler, I., P. Falkai, et al. (2009). "Disturbed functional connectivity within brain networks
subserving domain-specific subcomponents of working memory in schizophrenia:
Relation to performance and clinical symptoms." J Psychiatr Res.
157
Herbsman, T., D. Avery, et al. (2009). "More Lateral and Anterior Prefrontal Coil Location Is
Associated with Better Repetitive Transcranial Magnetic Stimulation Antidepressant
Response." Biol Psychiatry.
Herwig, U., F. Padberg, et al. (2001). "Transcranial magnetic stimulation in therapy studies:
examination of the reliability of "standard" coil positioning by neuronavigation." Biol
Psychiatry 50(1): 58-61.
Herwig, U., P. Satrapi, et al. (2003). "Using the international 10-20 EEG system for positioning
of transcranial magnetic stimulation." Brain Topogr 16(2): 95-9.
Heston, L. L. (1966). "Psychiatric disorders in foster home reared children of schizophrenic
mothers." Br J Psychiatry 112(489): 819-25.
Hoffman, R. E., N. N. Boutros, et al. (2000). "Transcranial magnetic stimulation and auditory
hallucinations in schizophrenia." Lancet 355(9209): 1073-5.
Hoffman, R. E., R. Gueorguieva, et al. (2005). "Temporoparietal transcranial magnetic
stimulation for auditory hallucinations: safety, efficacy and moderators in a fifty patient
sample." Biol Psychiatry 58(2): 97-104.
Hoffman, R. E., K. A. Hawkins, et al. (2003). "Transcranial magnetic stimulation of left
temporoparietal cortex and medication-resistant auditory hallucinations." Arch Gen
Psychiatry 60(1): 49-56.
Hogarty, G. E. (1993). "Prevention of relapse in chronic schizophrenic patients." J Clin
Psychiatry 54 Suppl: 18-23.
Holcomb, H. H., A. C. Lahti, et al. (2005). "Effects of noncompetitive NMDA receptor blockade
on anterior cingulate cerebral blood flow in volunteers with schizophrenia."
Neuropsychopharmacology 30(12): 2275-82.
Honey, G. D., E. T. Bullmore, et al. (2002). "De-coupling of cognitive performance and cerebral
functional response during working memory in schizophrenia." Schizophr Res 53(1-2):
45-56.
Honey, G. D., E. T. Bullmore, et al. (1999). "Differences in frontal cortical activation by a
working memory task after substitution of risperidone for typical antipsychotic drugs in
patients with schizophrenia." Proc Natl Acad Sci U S A 96(23): 13432-7.
Hong, L. E., A. Summerfelt, et al. (2004). "Evoked gamma band synchronization and the
liability for schizophrenia." Schizophr Res 70(2-3): 293-302.
Hong, L. E., A. Summerfelt, et al. (2004). "Gamma/beta oscillation and sensory gating deficit in
schizophrenia." Neuroreport 15(1): 155-9.
Horn, D. and M. Usher (1992). Oscillatory model of short-term memory. Advances in neural
information and processing systems. J. E. Moody, S. Hanson and R. Lippmann, Morgan
Kaufmann. 4.
158
Howard, M. W., D. S. Rizzuto, et al. (2003). "Gamma oscillations correlate with working
memory load in humans." Cereb Cortex 13(12): 1369-74.
Howes, O. D. and S. Kapur (2009). "The dopamine hypothesis of schizophrenia: version III--the
final common pathway." Schizophr Bull 35(3): 549-62.
Huber, T. J., U. Schneider, et al. (2003). "Gender differences in the effect of repetitive
transcranial magnetic stimulation in schizophrenia." Psychiatry Res 120(1): 103-5.
Ince, N. F., G. Pellizzer, et al. (2009). "Classification of schizophrenia with spectro-temporo-
spatial MEG patterns in working memory." Clin Neurophysiol 120(6): 1123-34.
Jablensky, A. (2000). "Epidemiology of schizophrenia: the global burden of disease and
disability." Eur Arch Psychiatry Clin Neurosci 250(6): 274-85.
Jakala, P., M. Riekkinen, et al. (1999). "Guanfacine, but not clonidine, improves planning and
working memory performance in humans." Neuropsychopharmacology 20(5): 460-70.
Jakala, P., J. Sirvio, et al. (1999). "Guanfacine and clonidine, alpha 2-agonists, improve paired
associates learning, but not delayed matching to sample, in humans."
Neuropsychopharmacology 20(2): 119-30.
Jandl, M., R. Bittner, et al. (2005). "Changes in negative symptoms and EEG in schizophrenic
patients after repetitive transcranial magnetic stimulation (rTMS): an open-label pilot
study." J Neural Transm 112(7): 955-67.
Jansma, J. M., N. F. Ramsey, et al. (2004). "Working memory capacity in schizophrenia: a
parametric fMRI study." Schizophr Res 68(2-3): 159-71.
Jasper, H. H. (1958). "The ten-twenty electrode system of the International Federation."
Electroencephalogr. Clin. Neurophysiol.: 370-375.
Jensen, O. (2006). "Maintenance of multiple working memory items by temporal segmentation."
Neuroscience 139(1): 237-49.
Jensen, O., J. Gelfand, et al. (2002). "Oscillations in the alpha band (9-12 Hz) increase with
memory load during retention in a short-term memory task." Cereb Cortex 12(8): 877-82.
Jensen, O. and J. E. Lisman (1998). "An oscillatory short-term memory buffer model can
account for data on the Sternberg task." J Neurosci 18(24): 10688-99.
Jensen, O. and C. D. Tesche (2002). "Frontal theta activity in humans increases with memory
load in a working memory task." Eur J Neurosci 15(8): 1395-9.
Jing, H. and M. Takigawa (2000). "Observation of EEG coherence after repetitive transcranial
magnetic stimulation." Clin Neurophysiol 111(9): 1620-31.
Johnson, J. L. (1972). "Glutamic acid as a synaptic transmitter in the nervous system. A review."
Brain Res 37(1): 1-19.
159
Jones, E. G. (1993). "GABAergic neurons and their role in cortical plasticity in primates." Cereb
Cortex 3(5): 361-72.
Jones, M. W., I. C. Kilpatrick, et al. (1986). "The agranular insular cortex: a site of unusually
high dopamine utilisation." Neurosci Lett 72(3): 330-4.
Jonides, J., E. H. Schumacher, et al. (1998). "The role of parietal cortex in verbal working
memory." J Neurosci 18(13): 5026-34.
Jonides, J., E. E. Smith, et al. (1993). "Spatial working memory in humans as revealed by PET."
Nature 363(6430): 623-5.
Joyce, J. N., A. Shane, et al. (1993). "Serotonin uptake sites and serotonin receptors are altered in
the limbic system of schizophrenics." Neuropsychopharmacology 8(4): 315-36.
Jung, S. H., J. E. Shin, et al. (2008). "Changes in motor cortical excitability induced by high-
frequency repetitive transcranial magnetic stimulation of different stimulation durations."
Clin Neurophysiol 119(1): 71-9.
Just, M. A. and P. A. Carpenter (1992). "A capacity theory of comprehension: individual
differences in working memory." Psychol Rev 99(1): 122-49.
Kahn, R. S., P. D. Harvey, et al. (1994). "Neuropsychological correlates of central monoamine
function in chronic schizophrenia: relationship between CSF metabolites and cognitive
function." Schizophr Res 11(3): 217-24.
Kaiser, J. and W. Lutzenberger (2005). "Human gamma-band activity: a window to cognitive
processing." Neuroreport 16(3): 207-11.
Kane, M. J., M. K. Bleckley, et al. (2001). "A controlled-attention view of working-memory
capacity." j Exp Psychol Gen 130: 169-183.
Kane, M. J., A. R. Conway, et al. (2007). "Working memory, attention control, and the N-back
task: a question of construct validity." J Exp Psychol Learn Mem Cogn 33(3): 615-22.
Kane, M. J. and R. W. Engle (2000). "Working-memory capacity, proactive interference, and
divided attention: limits on long-term memory retrieval." J Exp Psychol Learn Mem
Cogn 26(2): 336-58.
Kane, M. J. and R. W. Engle (2003). "Working-memory capacity and the control of attention: the
contributions of goal neglect, response competition, and task set to Stroop interference." J
Exp Psychol Gen 132(1): 47-70.
Kaplan, H. I., B. J. Sadock, et al. (1994). Kaplan and Sadock's synopsis of psychiatry :
behavioral sciences, clinical psychiatry. Baltimore, Williams & Wilkins.
Kargieman, L., N. Santana, et al. (2007). "Antipsychotic drugs reverse the disruption in
prefrontal cortex function produced by NMDA receptor blockade with phencyclidine."
Proc Natl Acad Sci U S A 104(37): 14843-8.
160
Kay, S. R., A. Fiszbein, et al. (1987). "The positive and negative syndrome scale (PANSS) for
schizophrenia." Schizophr Bull 13(2): 261-76.
Keck, M. E., T. Welt, et al. (2002). "Repetitive transcranial magnetic stimulation increases the
release of dopamine in the mesolimbic and mesostriatal system." Neuropharmacology
43(1): 101-9.
Keefe, R. S., R. M. Bilder, et al. (2007). "Neurocognitive effects of antipsychotic medications in
patients with chronic schizophrenia in the CATIE Trial." Arch Gen Psychiatry 64(6):
633-47.
Kendler, K. S. and C. D. Robinette (1983). "Schizophrenia in the National Academy of Sciences-
National Research Council Twin Registry: a 16-year update." Am J Psychiatry 140(12):
1551-63.
Kennedy, J. L., V. S. Basile, et al. (1999). "Chromosome 4 Workshop Summary: Sixth World
Congress on Psychiatric Genetics, Bonn, Germany, October 6-10, 1998." Am J Med
Genet 88(3): 224-8.
Khashan, A. S., K. M. Abel, et al. (2008). "Higher risk of offspring schizophrenia following
antenatal maternal exposure to severe adverse life events." Arch Gen Psychiatry 65(2):
146-52.
Khedr, E. M., J. C. Rothwell, et al. (2007). "Modulation of motor cortical excitability following
rapid-rate transcranial magnetic stimulation." Clin Neurophysiol 118(1): 140-5.
Khedr, E. M., J. C. Rothwell, et al. (2006). "Effect of daily repetitive transcranial magnetic
stimulation on motor performance in Parkinson's disease." Mov Disord 21(12): 2201-5.
Kievit, J. and H. G. Kuypers (1975). "Subcortical afferents to the frontal lobe in the rhesus
monkey studied by means of retrograde horseradish peroxidase transport." Brain Res
85(2): 261-6.
Kilman, V., M. C. van Rossum, et al. (2002). "Activity deprivation reduces miniature IPSC
amplitude by decreasing the number of postsynaptic GABA(A) receptors clustered at
neocortical synapses." J Neurosci 22(4): 1328-37.
Kindermann, S. S., G. G. Brown, et al. (2004). "Spatial working memory among middle-aged
and older patients with schizophrenia and volunteers using fMRI." Schizophr Res 68(2-
3): 203-16.
Klimesch, W. (1999). "EEG alpha and theta oscillations reflect cognitive and memory
performance: a review and analysis." Brain Res Brain Res Rev 29(2-3): 169-95.
Klimesch, W., M. Doppelmayr, et al. (1997). "Theta synchronization and alpha
desynchronization in a memory task." Psychophysiology 34(2): 169-76.
Klimesch, W., M. Doppelmayr, et al. (2001). "Episodic retrieval is reflected by a process specific
increase in human electroencephalographic theta activity." Neurosci Lett 302(1): 49-52.
161
Knable, M. B. and D. R. Weinberger (1997). "Dopamine, the prefrontal cortex and
schizophrenia." J Psychopharmacol 11(2): 123-31.
Knapp, M. (1997). "Costs of schizophrenia." Br J Psychiatry 171: 509-18.
Koos, T. and J. M. Tepper (1999). "Inhibitory control of neostriatal projection neurons by
GABAergic interneurons." Nat Neurosci 2(5): 467-72.
Krabbendam, L. and J. van Os (2005). "Schizophrenia and urbanicity: a major environmental
influence--conditional on genetic risk." Schizophr Bull 31(4): 795-9.
Krishnan, G. P., J. L. Vohs, et al. (2005). "Steady state visual evoked potential abnormalities in
schizophrenia." Clin Neurophysiol 116(3): 614-24.
Krnjevic, K. (1997). "Role of GABA in cerebral cortex." Can J Physiol Pharmacol 75(5): 439-
51.
Krystal, J. H., D. C. D'Souza, et al. (1999). "Interactive effects of subanesthetic ketamine and
haloperidol in healthy humans." Psychopharmacology (Berl) 145(2): 193-204.
Kulhara, P. and K. Chandiramani (1988). "Outcome of schizophrenia in India using various
diagnostic systems." Schizophr Res 1(5): 339-49.
Lahti, A. C., B. Koffel, et al. (1995). "Subanesthetic doses of ketamine stimulate psychosis in
schizophrenia." Neuropsychopharmacology 13(1): 9-19.
Lakatos, P., A. S. Shah, et al. (2005). "An oscillatory hierarchy controlling neuronal excitability
and stimulus processing in the auditory cortex." J Neurophysiol 94(3): 1904-11.
Lambert, T. J., D. Velakoulis, et al. (2003). "Medical comorbidity in schizophrenia." Med J Aust
178 Suppl: S67-70.
Lander, E. and L. Kruglyak (1995). "Genetic dissection of complex traits: guidelines for
interpreting and reporting linkage results." Nat Genet 11(3): 241-7.
Laruelle, M., R. M. Baldwin, et al. (1993). "SPECT imaging of dopamine and serotonin
transporters with [123I]beta-CIT: pharmacological characterization of brain uptake in
nonhuman primates." Synapse 13(4): 295-309.
Le Roux, N., M. Amar, et al. (2008). "Impaired GABAergic transmission disrupts normal
homeostatic plasticity in rat cortical networks." Eur J Neurosci 27(12): 3244-56.
Lee, J. and S. Park (2005). "Working memory impairments in schizophrenia: a meta-analysis." J
Abnorm Psychol 114(4): 599-611.
Lee, K. H., L. M. Williams, et al. (2003). ""Gamma (40 Hz) phase synchronicity" and symptom
dimensions in schizophrenia." Cogn Neuropsychiatry 8(1): 57-71.
162
Lee, S. W., J. G. Lee, et al. (2007). "A 12-week, double-blind, placebo-controlled trial of
galantamine adjunctive treatment to conventional antipsychotics for the cognitive
impairments in chronic schizophrenia." Int Clin Psychopharmacol 22(2): 63-8.
Leung, H. C., J. C. Gore, et al. (2002). "Sustained mnemonic response in the human middle
frontal gyrus during on-line storage of spatial memoranda." J Cogn Neurosci 14(4): 659-
71.
Leung, L. S. and B. Shen (2007). "GABAB receptor blockade enhances theta and gamma
rhythms in the hippocampus of behaving rats." Hippocampus 17(4): 281-91.
Levy, R. and P. S. Goldman-Rakic (2000). "Segregation of working memory functions within
the dorsolateral prefrontal cortex." Exp Brain Res 133(1): 23-32.
Lewis, D. A., R. Y. Cho, et al. (2008). "Subunit-selective modulation of GABA type A receptor
neurotransmission and cognition in schizophrenia." Am J Psychiatry 165(12): 1585-93.
Lewis, D. A., T. Hashimoto, et al. (2005). "Cortical inhibitory neurons and schizophrenia." Nat
Rev Neurosci 6(4): 312-24.
Lewis, D. A., J. N. Pierri, et al. (1999). "Altered GABA neurotransmission and prefrontal
cortical dysfunction in schizophrenia." Biol Psychiatry 46(5): 616-26.
Lieberman, J. A., J. M. Kane, et al. (1987). "Provocative tests with psychostimulant drugs in
schizophrenia." Psychopharmacology (Berl) 91(4): 415-33.
Lisman, J. E. and M. A. Idiart (1995). "Storage of 7 +/- 2 short-term memories in oscillatory
subcycles." Science 267(5203): 1512-5.
Little, J. T., T. A. Kimbrell, et al. (2000). "Cognitive effects of 1- and 20-hertz repetitive
transcranial magnetic stimulation in depression: preliminary report." Neuropsychiatry
Neuropsychol Behav Neurol 13(2): 119-24.
Liu, S. K., P. B. Fitzgerald, et al. (2009). "The relationship between cortical inhibition,
antipsychotic treatment, and the symptoms of schizophrenia." Biol Psychiatry 65(6): 503-
9.
Logothetis, N. K., J. Pauls, et al. (2001). "Neurophysiological investigation of the basis of the
fMRI signal." Nature 412(6843): 150-7.
Lorensen, W. E., Cline, H.E (1987). "Marching Cubes: A high resolution 3D surface
construction algorithm." Proceedings of SIGGRAPH 21(4): 163-169.
Luber, B., L. H. Kinnunen, et al. (2007). "Facilitation of performance in a working memory task
with rTMS stimulation of the precuneus: frequency- and time-dependent effects." Brain
Res 1128(1): 120-9.
Luby, E. D., B. D. Cohen, et al. (1959). "Study of a new schizophrenomimetic drug; sernyl."
AMA Arch Neurol Psychiatry 81(3): 363-9.
163
Luciana, M., E. D. Burgund, et al. (2001). "Effects of tryptophan loading on verbal, spatial and
affective working memory functions in healthy adults." J Psychopharmacol 15(4): 219-
30.
Luciana, M., P. F. Collins, et al. (1998). "Opposing roles for dopamine and serotonin in the
modulation of human spatial working memory functions." Cereb Cortex 8(3): 218-26.
Luck, S. J. and E. K. Vogel (1997). "The capacity of visual working memory for features and
conjunctions." Nature 390(6657): 279-81.
Lund, J. S. and D. A. Lewis (1993). "Local circuit neurons of developing and mature macaque
prefrontal cortex: Golgi and immunocytochemical characteristics." J Comp Neurol
328(2): 282-312.
Lutzenberger, W., F. Pulvermuller, et al. (1995). "Visual stimulation alters local 40-Hz responses
in humans: an EEG-study." Neurosci Lett 183(1-2): 39-42.
Lutzenberger, W., B. Ripper, et al. (2002). "Dynamics of gamma-band activity during an
audiospatial working memory task in humans." J Neurosci 22(13): 5630-8.
Magill, R. A., W. F. Waters, et al. (2003). "Effects of tyrosine, phentermine, caffeine D-
amphetamine, and placebo on cognitive and motor performance deficits during sleep
deprivation." Nutr Neurosci 6(4): 237-46.
Maier, W., S. Schwab, et al. (1994). "Absence of linkage between schizophrenia and the
dopamine D4 receptor gene." Psychiatry Res 53(1): 77-86.
Mainy, N., P. Kahane, et al. (2007). "Neural correlates of consolidation in working memory."
Hum Brain Mapp 28(3): 183-93.
Mann, E. O. and O. Paulsen (2007). "Role of GABAergic inhibition in hippocampal network
oscillations." Trends Neurosci 30(7): 343-9.
Manoach, D. S. (2003). "Prefrontal cortex dysfunction during working memory performance in
schizophrenia: reconciling discrepant findings." Schizophr Res 60(2-3): 285-98.
Manoach, D. S., R. L. Gollub, et al. (2000). "Schizophrenic subjects show aberrant fMRI
activation of dorsolateral prefrontal cortex and basal ganglia during working memory
performance." Biol Psychiatry 48(2): 99-109.
Marek, G. J., B. Behl, et al. "Glutamatergic (N-methyl-D-aspartate receptor) hypofrontality in
schizophrenia: too little juice or a miswired brain?" Mol Pharmacol 77(3): 317-26.
Markram, H., M. Toledo-Rodriguez, et al. (2004). "Interneurons of the neocortical inhibitory
system." Nat Rev Neurosci 5(10): 793-807.
Marrosu, F., F. Santoni, et al. (2006). "Beta and gamma range EEG power-spectrum correlation
with spiking discharges in DBA/2J mice absence model: role of GABA receptors."
Epilepsia 47(3): 489-94.
164
Martinkauppi, S., P. Rama, et al. (2000). "Working memory of auditory localization." Cereb
Cortex 10(9): 889-98.
Martis, B., D. Alam, et al. (2003). "Neurocognitive effects of repetitive transcranial magnetic
stimulation in severe major depression." Clin Neurophysiol 114(6): 1125-32.
Mattay, V. S., K. F. Berman, et al. (1996). "Dextroamphetamine enhances "neural network-
specific" physiological signals: a positron-emission tomography rCBF study." J Neurosci
16(15): 4816-22.
Mazziotta, J., A. Toga, et al. (2001). "A probabilistic atlas and reference system for the human
brain: International Consortium for Brain Mapping (ICBM)." Philos Trans R Soc Lond B
Biol Sci 356(1412): 1293-322.
McCormick, D. A. (1989). "GABA as an inhibitory neurotransmitter in human cerebral cortex."
J Neurophysiol 62(5): 1018-27.
McGrath, J. (1999). "Hypothesis: is low prenatal vitamin D a risk-modifying factor for
schizophrenia?" Schizophr Res 40(3): 173-7.
McGrath, J., S. Saha, et al. (2008). "Schizophrenia: a concise overview of incidence, prevalence,
and mortality." Epidemiol Rev 30: 67-76.
McGuffin, P., A. E. Farmer, et al. (1984). "Twin concordance for operationally defined
schizophrenia. Confirmation of familiality and heritability." Arch Gen Psychiatry 41(6):
541-5.
McGurk, S. R., E. W. Twamley, et al. (2007). "A meta-analysis of cognitive remediation in
schizophrenia." Am J Psychiatry 164(12): 1791-802.
McIntyre, D. C. and G. V. Goddard (1973). "Transfer, interference and spontaneous recovery of
convulsions kindled from the rat amygdala." Electroencephalogr Clin Neurophysiol
35(5): 533-43.
McKenna, P. J., D. Tamlyn, et al. (1990). "Amnesic syndrome in schizophrenia." Psychol Med
20(4): 967-72.
Meisenzahl, E. M., J. Scheuerecker, et al. (2006). "Effects of treatment with the atypical
neuroleptic quetiapine on working memory function: a functional MRI follow-up
investigation." Eur Arch Psychiatry Clin Neurosci 256(8): 522-31.
Melchitzky, D. S., S. R. Sesack, et al. (1998). "Synaptic targets of pyramidal neurons providing
intrinsic horizontal connections in monkey prefrontal cortex." J Comp Neurol 390(2):
211-24.
Meltzer, H. Y. and S. R. McGurk (1999). "The effects of clozapine, risperidone, and olanzapine
on cognitive function in schizophrenia." Schizophr Bull 25(2): 233-55.
165
Meltzer, J. A., H. P. Zaveri, et al. (2008). "Effects of working memory load on oscillatory power
in human intracranial EEG." Cereb Cortex 18(8): 1843-55.
Mendrek, A., K. A. Kiehl, et al. (2005). "Dysfunction of a distributed neural circuitry in
schizophrenia patients during a working-memory performance." Psychol Med 35(2): 187-
96.
Menzies, L., C. Ooi, et al. (2007). "Effects of gamma-aminobutyric acid-modulating drugs on
working memory and brain function in patients with schizophrenia." Arch Gen
Psychiatry 64(2): 156-67.
Millar, J. K., S. Christie, et al. (2001). "Genomic structure and localisation within a linkage
hotspot of Disrupted In Schizophrenia 1, a gene disrupted by a translocation segregating
with schizophrenia." Mol Psychiatry 6(2): 173-8.
Millar, J. K., J. C. Wilson-Annan, et al. (2000). "Disruption of two novel genes by a
translocation co-segregating with schizophrenia." Hum Mol Genet 9(9): 1415-23.
Mintzer, M. Z. and R. R. Griffiths (2003). "Triazolam-amphetamine interaction: dissociation of
effects on memory versus arousal." J Psychopharmacol 17(1): 17-29.
Miyake, A. and P. Shah (1999). Models of working memory: mechanisms of active maintenance
and executive control. Cambridge, Cambridge University Press.
Moghaddam, B., B. Adams, et al. (1997). "Activation of glutamatergic neurotransmission by
ketamine: a novel step in the pathway from NMDA receptor blockade to dopaminergic
and cognitive disruptions associated with the prefrontal cortex." J Neurosci 17(8): 2921-
7.
Mohn, A. R., R. R. Gainetdinov, et al. (1999). "Mice with reduced NMDA receptor expression
display behaviors related to schizophrenia." Cell 98(4): 427-36.
Moosmann, M., P. Ritter, et al. (2003). "Correlates of alpha rhythm in functional magnetic
resonance imaging and near infrared spectroscopy." Neuroimage 20(1): 145-58.
Mukamel, R., H. Gelbard, et al. (2005). "Coupling between neuronal firing, field potentials, and
FMRI in human auditory cortex." Science 309(5736): 951-4.
Muller, M. M., J. Bosch, et al. (1996). "Visually induced gamma-band responses in human
electroencephalographic activity--a link to animal studies." Exp Brain Res 112(1): 96-
102.
Murray, C. J. L. and A. D. Lopez (1996). "The global burden of disease." Boston: Harvard
University Press.
Naheed, M. and B. Green (2001). "Focus on clozapine." Curr Med Res Opin 17(3): 223-9.
166
Nakagawa, Y. and T. Takashima (1997). "The GABA(B) receptor antagonist CGP36742
attenuates the baclofen- and scopolamine-induced deficit in Morris water maze task in
rats." Brain Res 766(1-2): 101-6.
Nanko, S., R. Fukuda, et al. (1994). "Further evidence of no linkage between schizophrenia and
the dopamine D3 receptor gene locus." Am J Med Genet 54(3): 264-7.
Nasrallah, H. A. and D. J. Smeltzer (2002). Contemporary Diagnosis And Management Of The
Patient With Schizophrenia. Newtown, Pennsylvania, USA, Handbooks in Health Care
Company.
Neuper, C. and G. Pfurtscheller (2001). "Event-related dynamics of cortical rhythms: frequency-
specific features and functional correlates." Int J Psychophysiol 43(1): 41-58.
Neve, K. A., J. K. Seamans, et al. (2004). "Dopamine receptor signaling." J Recept Signal
Transduct Res 24(3): 165-205.
Newcomer, J. W. and J. H. Krystal (2001). "NMDA receptor regulation of memory and behavior
in humans." Hippocampus 11(5): 529-42.
Niessing, J., B. Ebisch, et al. (2005). "Hemodynamic signals correlate tightly with synchronized
gamma oscillations." Science 309(5736): 948-51.
Norman, G. R. and D. L. Streiner (2000). Biostatistics - The Bare Essentials. Hamilton, B.C.
Decker Inc.
Nuechterlein, K. H. and M. E. Dawson (1984). "Information processing and attentional
functioning in the developmental course of schizophrenic disorders." Schizophr Bull
10(2): 160-203.
Nurnberger, J. I., Jr. and T. Foroud (1999). "Chromosome 6 workshop report." Am J Med Genet
88(3): 233-8.
O'Connor, M. G., B. A. Jerskey, et al. (2005). "The effects of repetitive transcranial magnetic
stimulation (rTMS) on procedural memory and dysphoric mood in patients with major
depressive disorder." Cogn Behav Neurol 18(4): 223-7.
Ogawa, S., T. M. Lee, et al. (1990). "Brain magnetic resonance imaging with contrast dependent
on blood oxygenation." Proc Natl Acad Sci U S A 87(24): 9868-72.
Ohaeri, J. U. (1993). "Long-term outcome of treated schizophrenia in a Nigerian cohort.
Retrospective analysis of 7-year follow-ups." J Nerv Ment Dis 181(8): 514-6.
Ohuoha, D. C., T. M. Hyde, et al. (1993). "The role of serotonin in schizophrenia: an overview
of the nomenclature, distribution and alterations of serotonin receptors in the central
nervous system." Psychopharmacology (Berl) 112(1 Suppl): S5-15.
Okamura, H., H. Jing, et al. (2001). "EEG modification induced by repetitive transcranial
magnetic stimulation." J Clin Neurophysiol 18(4): 318-25.
167
Oldfield, R. C. (1971). "The assessment and analysis of handedness: the Edinburgh inventory."
Neuropsychologia 9(1): 97-113.
Oltmanns, T. F., J. Ohayon, et al. (1978). "The effect of antipsychotic medication and diagnostic
criteria on distractibility in schizophrenia." J Psychiatr Res 14(1-4): 81-91.
Owen, A. M. (1997). "The functional organization of working memory processes within human
lateral frontal cortex: the contribution of functional neuroimaging." Eur J Neurosci 9(7):
1329-39.
Owen, A. M., K. M. McMillan, et al. (2005). "N-back working memory paradigm: a meta-
analysis of normative functional neuroimaging studies." Hum Brain Mapp 25(1): 46-59.
Owen, M. J. (2000). "Molecular genetic studies of schizophrenia." Brain Res Brain Res Rev
31(2-3): 179-86.
Owen, M. J. and N. Craddock (1996). "Modern molecular genetic approaches to complex traits:
implications for psychiatric disorders." Mol Psychiatry 1(1): 21-6.
Pantev, C. (1995). "Evoked and induced gamma-band activity of the human cortex." Brain
Topogr 7(4): 321-30.
Park, S., P. S. Holzman, et al. (1995). "Spatial working memory deficits in the relatives of
schizophrenic patients." Arch Gen Psychiatry 52(10): 821-8.
Park, S., J. Puschel, et al. (1999). "Spatial working memory deficits and clinical symptoms in
schizophrenia: a 4-month follow-up study." Biol Psychiatry 46(3): 392-400.
Pascual-Leone, A., C. M. Houser, et al. (1993). "Safety of rapid-rate transcranial magnetic
stimulation in normal volunteers." Electroencephalogr Clin Neurophysiol 89(2): 120-30.
Pascual-Leone, A., D. S. Manoach, et al. (2002). "Motor cortical excitability in schizophrenia."
Biol Psychiatry 52(1): 24-31.
Pascual-Leone, A., B. Rubio, et al. (1996). "Rapid-rate transcranial magnetic stimulation of left
dorsolateral prefrontal cortex in drug-resistant depression." Lancet 348(9022): 233-7.
Paterson, A. D. (1999). "Sixth World Congress of Psychiatric Genetics X Chromosome
Workshop." Am J Med Genet 88(3): 279-86.
Penttonen, M. and G. Buzsaki (2003). "Natural logarithmic relationship between brain
oscillators." Thalamus and related systems 2(2): 145-152.
Penttonen, M. and G. Buzsaki (2003). "Natural logarithmic relationship between brain
oscillators." Thalamus Relat Syst 2: 145-152.
Perlstein, W. M., C. S. Carter, et al. (2001). "Relation of prefrontal cortex dysfunction to
working memory and symptoms in schizophrenia." Am J Psychiatry 158(7): 1105-13.
168
Pesaran, B., J. S. Pezaris, et al. (2002). "Temporal structure in neuronal activity during working
memory in macaque parietal cortex." Nat Neurosci 5(8): 805-11.
Peters, A. (1984). Cerebral Cortex. Plenum, New York.
Petrides, M. (1995). "Impairments on nonspatial self-ordered and externally ordered working
memory tasks after lesions of the mid-dorsal part of the lateral frontal cortex in the
monkey." J Neurosci 15(1 Pt 1): 359-75.
Petrides, M. (2000). "The role of the mid-dorsolateral prefrontal cortex in working memory."
Exp Brain Res 133(1): 44-54.
Petrides, M. and B. Milner (1982). "Deficits on subject-ordered tasks after frontal- and temporal-
lobe lesions in man." Neuropsychologia 20(3): 249-62.
Petrides, M. and D. N. Pandya (1984). "Projections to the frontal cortex from the posterior
parietal region in the rhesus monkey." J Comp Neurol 228(1): 105-16.
Pfurtscheller, G. and F. H. Lopes da Silva (1999). "Event-related EEG/MEG synchronization and
desynchronization: basic principles." Clin Neurophysiol 110(11): 1842-57.
Pigeau, R., P. Naitoh, et al. (1995). "Modafinil, d-amphetamine and placebo during 64 hours of
sustained mental work. I. Effects on mood, fatigue, cognitive performance and body
temperature." J Sleep Res 4(4): 212-228.
Post, A. and M. E. Keck (2001). "Transcranial magnetic stimulation as a therapeutic tool in
psychiatry: what do we know about the neurobiological mechanisms?" J Psychiatr Res
35(4): 193-215.
Postle, B. R. (2006). "Working memory as an emergent property of the mind and brain."
Neuroscience 139(1): 23-38.
Postle, B. R. and E. Feredoes (2009). "Stronger inference with direct manipulation of brain
function." Cortex 46(1): 121-3.
Poulet, E., J. Brunelin, et al. (2005). "Slow transcranial magnetic stimulation can rapidly reduce
resistant auditory hallucinations in schizophrenia." Biol Psychiatry 57(2): 188-91.
Preston, G., E. Anderson, et al. (2009). "Effects of 10 Hz rTMS on the neural efficiency of
working memory." J Cogn Neurosci 22(3): 447-56.
Pucak, M. L., J. B. Levitt, et al. (1996). "Patterns of intrinsic and associational circuitry in
monkey prefrontal cortex." J Comp Neurol 376(4): 614-30.
Raghavachari, S., M. J. Kahana, et al. (2001). "Gating of human theta oscillations by a working
memory task." J Neurosci 21(9): 3175-83.
169
Ranganath, C., M. X. Cohen, et al. (2004). "Inferior temporal, prefrontal, and hippocampal
contributions to visual working memory maintenance and associative memory retrieval."
J Neurosci 24(16): 3917-25.
Rao, S. G., G. V. Williams, et al. (2000). "Destruction and creation of spatial tuning by
disinhibition: GABA(A) blockade of prefrontal cortical neurons engaged by working
memory." J Neurosci 20(1): 485-94.
Ravindranathan, A., H. Coon, et al. (1994). "Linkage analysis between schizophrenia and a
microsatellite polymorphism for the D5 dopamine receptor gene." Psychiatr Genet 4(2):
77-80.
Raymond, J. R., Y. V. Mukhin, et al. (2001). "Multiplicity of mechanisms of serotonin receptor
signal transduction." Pharmacol Ther 92(2-3): 179-212.
Reichenberg, A., M. Weiser, et al. (2006). "Premorbid intellectual functioning and risk of
schizophrenia and spectrum disorders." J Clin Exp Neuropsychol 28(2): 193-207.
Riley, B. P. and P. McGuffin (2000). "Linkage and associated studies of schizophrenia." Am J
Med Genet 97(1): 23-44.
Rizzuto, D. S., J. R. Madsen, et al. (2003). "Reset of human neocortical oscillations during a
working memory task." Proc Natl Acad Sci U S A 100(13): 7931-6.
Roberts, E. (1972). "Prospects for research on schizophrenia. An hypotheses suggesting that
there is a defect in the GABA system in schizophrenia." Neurosci Res Program Bull
10(4): 468-82.
Roberts, E. and S. Frankel (1950). "gamma-Aminobutyric acid in brain: its formation from
glutamic acid." J Biol Chem 187(1): 55-63.
Romanides, A. J., P. Duffy, et al. (1999). "Glutamatergic and dopaminergic afferents to the
prefrontal cortex regulate spatial working memory in rats." Neuroscience 92(1): 97-106.
Romanski, L. M. (2004). "Domain specificity in the primate prefrontal cortex." Cogn Affect
Behav Neurosci 4(4): 421-9.
Roopun, A. K., M. A. Kramer, et al. (2008). "Temporal Interactions between Cortical Rhythms."
Front Neurosci 2(2): 145-54.
Rosanova, M., A. Casali, et al. (2009). "Natural frequencies of human corticothalamic circuits." J
Neurosci 29(24): 7679-85.
Rossi, S., S. F. Cappa, et al. (2001). "Prefrontal [correction of Prefontal] cortex in long-term
memory: an "interference" approach using magnetic stimulation." Nat Neurosci 4(9):
948-52.
170
Rusjan, P. M., M. S. Barr, et al. (2010). "Optimal transcranial magnetic stimulation coil
placement for targeting the dorsolateral prefrontal cortex using novel magnetic resonance
image-guided neuronavigation." Hum Brain Mapp 31(11): 1643-1652.
Sachdev, P., C. Loo, et al. (2005). "Transcranial magnetic stimulation for the deficit syndrome of
schizophrenia: a pilot investigation." Psychiatry Clin Neurosci 59(3): 354-7.
Sackeim, H. A., S. Portnoy, et al. (1986). "Cognitive consequences of low-dosage
electroconvulsive therapy." Ann N Y Acad Sci 462: 326-40.
Sakai, K., J. B. Rowe, et al. (2002). "Active maintenance in prefrontal area 46 creates distractor-
resistant memory." Nat Neurosci 5(5): 479-84.
Saliba, R. S., Z. Gu, et al. (2009). "Blocking L-type voltage-gated Ca2+ channels with
dihydropyridines reduces gamma-aminobutyric acid type A receptor expression and
synaptic inhibition." J Biol Chem 284(47): 32544-50.
Saliba, R. S., G. Michels, et al. (2007). "Activity-dependent ubiquitination of GABA(A)
receptors regulates their accumulation at synaptic sites." J Neurosci 27(48): 13341-51.
Sandrini, M., P. M. Rossini, et al. (2008). "Lateralized contribution of prefrontal cortex in
controlling task-irrelevant information during verbal and spatial working memory tasks:
rTMS evidence." Neuropsychologia 46(7): 2056-63.
Sanger, T. D., R. R. Garg, et al. (2001). "Interactions between two different inhibitory systems in
the human motor cortex." J Physiol 530(Pt 2): 307-17.
Sarnthein, J., H. Petsche, et al. (1998). "Synchronization between prefrontal and posterior
association cortex during human working memory." Proc Natl Acad Sci U S A 95(12):
7092-6.
Sawaguchi, T. and P. S. Goldman-Rakic (1994). "The role of D1-dopamine receptor in working
memory: local injections of dopamine antagonists into the prefrontal cortex of rhesus
monkeys performing an oculomotor delayed-response task." J Neurophysiol 71(2): 515-
28.
Sawaguchi, T., M. Matsumura, et al. (1989). "Delayed response deficits produced by local
injection of bicuculline into the dorsolateral prefrontal cortex in Japanese macaque
monkeys." Exp Brain Res 75(3): 457-69.
Schack, B., N. Vath, et al. (2002). "Phase-coupling of theta-gamma EEG rhythms during short-
term memory processing." Int J Psychophysiol 44(2): 143-63.
Schlosser, R. G., I. Nenadic, et al. (2007). "White matter abnormalities and brain activation in
schizophrenia: a combined DTI and fMRI study." Schizophr Res 89(1-3): 1-11.
Schmiedt, C., A. Brand, et al. (2005). "Event-related theta oscillations during working memory
tasks in patients with schizophrenia and healthy controls." Brain Res Cogn Brain Res
25(3): 936-47.
171
Schubert, M. H., K. A. Young, et al. (2006). "Galantamine improves cognition in schizophrenic
patients stabilized on risperidone." Biol Psychiatry 60(6): 530-3.
Schultze-Rauchenbach, S. C., U. Harms, et al. (2005). "Distinctive neurocognitive effects of
repetitive transcranial magnetic stimulation and electroconvulsive therapy in major
depression." Brit J Psych 186: 410-416.
Schwab, S. G. and D. B. Wildenauer (1999). "Chromosome 22 workshop report." Am J Med
Genet 88(3): 276-8.
Seeman, P. and T. Lee (1975). "Antipsychotic drugs: direct correlation between clinical potency
and presynaptic action on dopamine neurons." Science 188(4194): 1217-9.
Sejnowski, T. J. (1977). "Statistical constraints on synaptic plasticity." J Theor Biol 69(2): 385-9.
Shackman, A. J. (2010). "The potentially deleterious impact of muscle activity on gamma band
references." Neuropsychopharmacology 35: 847.
Sharma, T., C. Hughes, et al. (2003). "Cognitive effects of olanzapine and clozapine treatment in
chronic schizophrenia." Psychopharmacology (Berl) 169(3-4): 398-403.
Shelton, J. T., R. L. Metzger, et al. (2007). "A group-administered lag task as a measure of
working memory." Behav Res Methods 39(3): 482-93.
Siebner, H. R., N. Lang, et al. (2004). "Preconditioning of low-frequency repetitive transcranial
magnetic stimulation with transcranial direct current stimulation: evidence for
homeostatic plasticity in the human motor cortex." J Neurosci 24(13): 3379-85.
Silver, H., P. Feldman, et al. (2003). "Working memory deficit as a core neuropsychological
dysfunction in schizophrenia." Am J Psychiatry 160(10): 1809-16.
Simpson, M. D., D. I. Lubman, et al. (1996). "Autoradiography with [3H]8-OH-DPAT reveals
increases in 5-HT(1A) receptors in ventral prefrontal cortex in schizophrenia." Biol
Psychiatry 39(11): 919-28.
Singer, W. (1999). "Neuronal synchrony: a versatile code for the definition of relations?" Neuron
24(1): 49-65, 111-25.
Smith, E. E. and J. Jonides (1999). "Storage and executive processes in the frontal lobes."
Science 283(5408): 1657-61.
Snitz, B. E., A. MacDonald, 3rd, et al. (2005). "Lateral and medial hypofrontality in first-episode
schizophrenia: functional activity in a medication-naive state and effects of short-term
atypical antipsychotic treatment." Am J Psychiatry 162(12): 2322-9.
Snitz, B. E., A. W. Macdonald, 3rd, et al. (2006). "Cognitive deficits in unaffected first-degree
relatives of schizophrenia patients: a meta-analytic review of putative endophenotypes."
Schizophr Bull 32(1): 179-94.
172
Sohal, V. S., F. Zhang, et al. (2009). "Parvalbumin neurons and gamma rhythms enhance cortical
circuit performance." Nature 459(7247): 698-702.
Somogyi, P. and A. Cowey (1981). "Combined Golgi and electron microscopic study on the
synapses formed by double bouquet cells in the visual cortex of the cat and monkey." J
Comp Neurol 195(4): 547-66.
Somogyi, P., G. Tamas, et al. (1998). "Salient features of synaptic organisation in the cerebral
cortex." Brain Res Brain Res Rev 26(2-3): 113-35.
Sparing, R., F. M. Mottaghy, et al. (2001). "Repetitive transcranial magnetic stimulation effects
on language function depend on the stimulation parameters." J Clin Neurophysiol 18(4):
326-30.
Spencer, K. M., M. A. Niznikiewicz, et al. (2008). "Sensory-evoked gamma oscillations in
chronic schizophrenia." Biol Psychiatry 63(8): 744-7.
Spitzer, R. L. (1994). Diagnostic and Statistical Manual of Mental Disorders, 4th Edition.
Washington, D.C., American Psychiatric Association.
Squire, L. R. (1982). "Memory and electroconvulsive therapy." Am J Psychiatry 139(9): 1221.
Squires, R. F. and E. Saederup (2000). "Additivities of compounds that increase the numbers of
high affinity [3H]muscimol binding sites by different amounts define more than 9
GABA(A) receptor complexes in rat forebrain: implications for schizophrenia and
clozapine research." Neurochem Res 25(12): 1587-601.
Srinivasan, R., W. R. Winter, et al. (2006). "Source analysis of EEG oscillations using high-
resolution EEG and MEG." Prog Brain Res 159: 29-42.
Stephane, M., N. F. Ince, et al. (2008). "Temporospatial characterization of brain oscillations
(TSCBO) associated with subprocesses of verbal working memory in schizophrenia."
Clin EEG Neurosci 39(4): 194-202.
Strafella, A. P., T. Paus, et al. (2001). "Repetitive transcranial magnetic stimulation of the human
prefrontal cortex induces dopamine release in the caudate nucleus." J Neurosci 21(15):
RC157.
Sumiyoshi, T., C. A. Stockmeier, et al. (1996). "Serotonin1A receptors are increased in
postmortem prefrontal cortex in schizophrenia." Brain Res 708(1-2): 209-14.
Swanwick, C. C., N. R. Murthy, et al. (2006). "Activity-dependent scaling of GABAergic
synapse strength is regulated by brain-derived neurotrophic factor." Mol Cell Neurosci
31(3): 481-92.
Swerdlow, N. R. and G. F. Koob (1987). "Dopamine, schizophrenia, mania, depression: Toward
a unified hypothesis of cortico-striatal-pallido-thalamic function." Behaviour and Brain
Sciences 10: 197-245.
173
Takano, B., A. Drzezga, et al. (2004). "Short-term modulation of regional excitability and blood
flow in human motor cortex following rapid-rate transcranial magnetic stimulation."
Neuroimage 23(3): 849-59.
Tallon-Baudry, C. and O. Bertrand (1999). "Oscillatory gamma activity in humans and its role in
object representation." Trends Cogn Sci 3(4): 151-162.
Tallon-Baudry, C., O. Bertrand, et al. (1996). "Stimulus specificity of phase-locked and non-
phase-locked 40 Hz visual responses in human." J Neurosci 16(13): 4240-9.
Tallon-Baudry, C., O. Bertrand, et al. (2001). "Oscillatory synchrony between human extrastriate
areas during visual short-term memory maintenance." J Neurosci 21(20): RC177.
Tallon-Baudry, C., O. Bertrand, et al. (1998). "Induced gamma-band activity during the delay of
a visual short-term memory task in humans." J Neurosci 18(11): 4244-54.
Tallon-Baudry, C., A. Kreiter, et al. (1999). "Sustained and transient oscillatory responses in the
gamma and beta bands in a visual short-term memory task in humans." Vis Neurosci
16(3): 449-59.
Tamas, G., E. H. Buhl, et al. (2000). "Proximally targeted GABAergic synapses and gap
junctions synchronize cortical interneurons." Nat Neurosci 3(4): 366-71.
Tamminga, C. A. (1998). "Schizophrenia and glutamatergic transmission." Crit Rev Neurobiol
12(1-2): 21-36.
Tan, H. Y., W. C. Choo, et al. (2005). "fMRI study of maintenance and manipulation processes
within working memory in first-episode schizophrenia." Am J Psychiatry 162(10): 1849-
58.
Ter-Pogossian, M. M., M. E. Phelps, et al. (1975). "A positron-emission transaxial tomograph
for nuclear imaging (PETT)." Radiology 114(1): 89-98.
Tesche, C. D. and J. Karhu (2000). "Theta oscillations index human hippocampal activation
during a working memory task." Proc Natl Acad Sci U S A 97(2): 919-24.
Thompson, P. and A. W. Toga (1996). "A surface-based technique for warping three-
dimensional images of the brain." IEEE Trans Med Imaging 15(4): 402-17.
Thomson, A. M. (2000). "Neurotransmission: Chemical and electrical interneuron coupling."
Curr Biol 10(3): R110-2.
Traub, R. D., H. Michelson-Law, et al. (2004). "Gap junctions, fast oscillations and the initiation
of seizures." Adv Exp Med Biol 548: 110-22.
Traub, R. D., M. A. Whittington, et al. (1999). "On the mechanism of the gamma --> beta
frequency shift in neuronal oscillations induced in rat hippocampal slices by tetanic
stimulation." J Neurosci 19(3): 1088-105.
174
Traub, R. D., M. A. Whittington, et al. (1996). "Analysis of gamma rhythms in the rat
hippocampus in vitro and in vivo." J Physiol 493 ( Pt 2): 471-84.
Trippe, J., A. Mix, et al. (2009). "Theta burst and conventional low-frequency rTMS
differentially affect GABAergic neurotransmission in the rat cortex." Exp Brain Res.
Uhlhaas, P. J., D. E. Linden, et al. (2006). "Dysfunctional long-range coordination of neural
activity during Gestalt perception in schizophrenia." J Neurosci 26(31): 8168-75.
Unsworth, N., J. C. Schrock, et al. (2004). "Working memory capacity and the antisaccade task:
individual differences in voluntary saccade control." J Exp Psychol Learn Mem Cogn
30(6): 1302-21.
Van Broeckhoven, C. and G. Verheyen (1999). "Report of the chromosome 18 workshop." Am J
Med Genet 88(3): 263-70.
Van Snellenberg, J. X., I. J. Torres, et al. (2006). "Functional neuroimaging of working memory
in schizophrenia: task performance as a moderating variable." Neuropsychology 20(5):
497-510.
Veling, W., J. P. Mackenbach, et al. (2008). "Cannabis use and genetic predisposition for
schizophrenia: a case-control study." Psychol Med 38(9): 1251-6.
Ventura, J., G. S. Hellemann, et al. (2009). "Symptoms as mediators of the relationship between
neurocognition and functional outcome in schizophrenia: a meta-analysis." Schizophr
Res 113(2-3): 189-99.
Vidal, J. R., M. Chaumon, et al. (2006). "Visual grouping and the focusing of attention induce
gamma-band oscillations at different frequencies in human magnetoencephalogram
signals." J Cogn Neurosci 18(11): 1850-62.
Villalta-Gil, V., M. Vilaplana, et al. (2006). "Neurocognitive performance and negative
symptoms: are they equal in explaining disability in schizophrenia outpatients?"
Schizophr Res 87(1-3): 246-53.
Volk, D. W., J. N. Pierri, et al. (2002). "Reciprocal alterations in pre- and postsynaptic inhibitory
markers at chandelier cell inputs to pyramidal neurons in schizophrenia." Cereb Cortex
12(10): 1063-70.
von Bohlen, R. (2006). Neurotransmitters and neuromodulators. Weinheim, Germany, Wiley-
VCH.
von Stein, A., P. Rappelsberger, et al. (1999). "Synchronization between temporal and parietal
cortex during multimodal object processing in man." Cereb Cortex 9(2): 137-50.
Vreugdenhil, M., J. G. Jefferys, et al. (2003). "Parvalbumin-deficiency facilitates repetitive
IPSCs and gamma oscillations in the hippocampus." J Neurophysiol 89(3): 1414-22.
175
Wager, T. D. and E. E. Smith (2003). "Neuroimaging studies of working memory: a meta-
analysis." Cogn Affect Behav Neurosci 3(4): 255-74.
Walker, E. F. (1994). "Developmentally moderated expressions of the neuropathology
underlying schizophrenia." Schizophr Bull 20(3): 453-80.
Walter, H., N. Vasic, et al. (2007). "Working memory dysfunction in schizophrenia compared to
healthy controls and patients with depression: evidence from event-related fMRI."
Neuroimage 35(4): 1551-61.
Walter, H., A. P. Wunderlich, et al. (2003). "No hypofrontality, but absence of prefrontal
lateralization comparing verbal and spatial working memory in schizophrenia."
Schizophr Res 61(2-3): 175-84.
Wang, X. J. and G. Buzsaki (1996). "Gamma oscillation by synaptic inhibition in a hippocampal
interneuronal network model." J Neurosci 16(20): 6402-13.
Weickert, T. W., T. E. Goldberg, et al. (2000). "Cognitive impairments in patients with
schizophrenia displaying preserved and compromised intellect." Arch Gen Psychiatry
57(9): 907-13.
Weinberger, D. R., K. F. Berman, et al. (1988). "Mesocortical dopaminergic function and human
cognition." Ann N Y Acad Sci 537: 330-8.
Weinberger, D. R., K. F. Berman, et al. (1986). "Physiologic dysfunction of dorsolateral
prefrontal cortex in schizophrenia. I. Regional cerebral blood flow evidence." Arch Gen
Psychiatry 43(2): 114-24.
Weiner, R. D., H. J. Rogers, et al. (1986). "Effects of stimulus parameters on cognitive side
effects." Ann N Y Acad Sci 462: 315-25.
Wender, P. H., D. Rosenthal, et al. (1977). "Schizophrenics' adopting parents. Psychiatric status."
Arch Gen Psychiatry 34(7): 777-84.
Wheeler, M. E. and A. M. Treisman (2002). "Binding in short-term memory." j Exp Psychol Gen
131: 48-64.
Whittington, M. A., I. M. Stanford, et al. (1997). "Spatiotemporal patterns of gamma frequency
oscillations tetanically induced in the rat hippocampal slice." J Physiol 502 ( Pt 3): 591-
607.
Whittington, M. A., R. D. Traub, et al. (1995). "Synchronized oscillations in interneuron
networks driven by metabotropic glutamate receptor activation." Nature 373(6515): 612-
5.
Wildenauer, D. B. and S. G. Schwab (1999). "Chromosomes 8 and 10 workshop." Am J Med
Genet 88(3): 239-43.
176
Wilhelm, J. C. and P. Wenner (2008). "GABAA transmission is a critical step in the process of
triggering homeostatic increases in quantal amplitude." Proc Natl Acad Sci U S A
105(32): 11412-7.
Williams, G. V. and P. S. Goldman-Rakic (1995). "Modulation of memory fields by dopamine
D1 receptors in prefrontal cortex." Nature 376(6541): 572-5.
Williams, J., G. Spurlock, et al. (1996). "Association between schizophrenia and T102C
polymorphism of the 5-hydroxytryptamine type 2a-receptor gene. European Multicentre
Association Study of Schizophrenia (EMASS) Group." Lancet 347(9011): 1294-6.
Wilson, F. A., S. P. O'Scalaidhe, et al. (1994). "Functional synergism between putative gamma-
aminobutyrate-containing neurons and pyramidal neurons in prefrontal cortex." Proc Natl
Acad Sci U S A 91(9): 4009-13.
Wohl, M. and P. Gorwood (2007). "Paternal ages below or above 35 years old are associated
with a different risk of schizophrenia in the offspring." Eur Psychiatry 22(1): 22-6.
Wolf, J., F. Janssen, et al. (2007). "A prospective, multicentre, open-label study of aripiprazole in
the management of patients with schizophrenia in psychiatric practice in Europe: Broad
Effectiveness Trial with Aripiprazole in Europe (EU-BETA)." Curr Med Res Opin
23(10): 2313-23.
Wong, A. H. and H. H. Van Tol (2003). "Schizophrenia: from phenomenology to neurobiology."
Neurosci Biobehav Rev 27(3): 269-306.
Woods, S. W. (2003). "Chlorpromazine equivalent doses for the newer atypical antipsychotics."
J Clin Psychiatry 64(6): 663-7.
Woodward, N. D., S. E. Purdon, et al. (2005). "A meta-analysis of neuropsychological change to
clozapine, olanzapine, quetiapine, and risperidone in schizophrenia." Int J
Neuropsychopharmacol 8(3): 457-72.
Wooley, D. W. and E. Shaw (1954). "A biological and pharmacological suggestion about certain
mental disorder." Proc Natl Acad Sci U S A 40: 228–231.
Wu, E. Q., H. G. Birnbaum, et al. (2002). "The economic burden of schizophrenia in the United
States in 2002." J Clin Psychiatry 66(9): 1122-1129.
Yuval-Greenberg, S., O. Tomer, et al. (2008). "Transient induced gamma-band response in EEG
as a manifestation of miniature saccades." Neuron 58(3): 429-41.
Ziemann, U., S. Lonnecker, et al. (1996). "The effect of lorazepam on the motor cortical
excitability in man." Exp Brain Res 109(1): 127-35.
177
7 Appendix A: Optimal TMS Coil Placement for Targeting the DLPFC Using Novel MRI-Guided Neuronavigation
7.1 Abstract
The dorsolateral prefrontal cortex (DLPFC) has been implicated in the pathophysiology
of several psychiatric illnesses including major depressive disorder and schizophrenia. In this
regard, the DLPFC has been targeted in many repetitive transcranial magnetic stimulation
(rTMS) studies as a form of treatment to those patients who are resistant to medications. The ‗5-
cm method‘ and the ‗10-20 method‘ for positioning the transcranial magnetic stimulation (TMS)
coil over DLPFC have been scrutinized due to poor targeting accuracies attributed to inter-
subject variability. We evaluated the accuracy of such methods to localize the DLPFC on the
scalp in 15 healthy subjects and compared them to our novel neuronavigational method, which
first estimates the DLPFC position in the cortex based on a standard template and then
determines the most appropriate position on the scalp in which to place the TMS coil. Our
neuronavigational method yielded a scalp position for the left DLPFC between electrodes F3 and
F5 in standard space and was closest to electrode F5 in individual space. Further, we found that
there was significantly less inter-subject variability using our neuronavigational method for
localizing the DLPFC on the scalp compared to the ‗5-cm method‘ and the ‗10-20 method‘. Our
findings also suggest that the ‗10-20 method‘ is superior to the ‗5-cm method‘ in reducing inter-
subject variability and that electrode F5 should be the stimulation location of choice when MRI
co-registration is not available.
178
7.2 Introduction
The abnormal functioning of the dorsolateral prefrontal cortex (DLPFC) has been
implicated in the pathophysiology of several neuropsychiatric disorders including major
depressive disorder (MDD) and schizophrenia. Moreover, such deficits in DLPFC functioning
have been related to symptom severity and serves as a target for repetitive transcranial magnetic
stimulation (rTMS) as an alternative treatment. Given the promising success of rTMS over the
DLPFC in alleviating symptoms associated with MDD and schizophrenia, advancement in
improving the localization of this brain region is invaluable to optimizing rTMS as a potential
treatment.
Over the past decade the ‗5-cm method‘ has been routinely used for targeting the DLPFC
in transcranial magnetic stimulation (TMS) studies. With this method, the DLPFC is localized
by stimulating the motor cortex and recording motor evoked potentials in the contralateral hand
muscle (i.e., abductor pollicis brevis; APB), and then measuring 5 cm anterior from this position
along a parasaggital line (George, Wassermann et al. 1995; Pascual-Leone, Rubio et al. 1996).
Although the ‗5-cm method‘ is easy to perform without the cost of expensive neuronavigational
techniques, it has been criticized for its failure to take into account cortex morphology and often
yield positions that fall too short (Herwig, Padberg et al. 2001). For example, an MRI-based
neuronavagational study reported that the ‗5-cm method‘ was found to correspond to Brodman‘s
area (BA) 6 (premotor cortex) or 8 (frontal eye field) in 15/22 subjects, while a position within
BA 9 was found in the other 7 subjects. Such inter-individual variation for identifying the
DLPFC through the ‗5-cm method‘ is thought to be a result of both variations in the distance
between the motor areas and the DLPFC, and also low inter-rater reliability in determining this
distance. Most importantly, such variability may also account for inconsistent therapeutic results
179
when using repetitive TMS (rTMS) over the DLPFC to treat patients with MDD. For example, a
recent study examined DLPFC location variability of the ‗5-cm method‘ with rTMS anti-
depressive effects and found significantly reduced depression scores in individuals with DLPFC
positions that were more lateral and anterior (Herbsman, Avery et al. 2009). Furthermore, our
group showed that targeted rTMS to DLPFC functional coordinates guided by a meta-analysis on
neuorimaging studies on working memory in MDD patients (Fitzgerald, Oxley et al. 2006)
resulted in greater anti-depressive effects compared to the ‗5-cm method‘ (Fitzgerald, Hoy et al.
2009). Together these studies demonstrate the pitfalls of the ‗5-cm method‘ and provide clear
support for the need for better localization techniques that are less susceptible to inter-subject
and inter-rater variability to optimize rTMS as a treatment for MDD.
The international 10-20 system (Jasper 1958) that is typically used in
electroencephalography (EEG) electrode placement has proven useful in linking external scalp
locations to underlying cortex. To this end, previous studies have used the 10-20 system to
target the left and right DLPFC by placing the TMS coil over the F3 and F4 electrode,
respectively (Gerloff, Corwell et al. 1997; Rossi, Cappa et al. 2001). Herwig and colleagues
(2003) provided further support in ascribing F3 and F4 from the 10-20 system to the underlying
DLPFC cortical area (Herwig, Satrapi et al. 2003). In their study, 21 subjects were co-registered
with an EEG cap and dynamic reference frame mounted on their head. With their
neuronavigational technique based on frameless stereotaxy, a probe was placed over electrodes
F3 and F4 and its position projected 15 mm down to the underlying cortex to provide X, Y, and
Z coordinates in individual space. The authors then used a Talairach atlas to relate individual
coordinates to Talairach coordinates (x, y, z) = -37, 27, 44 for the left DLPFC, which
corresponded to the dorsal and superior edge of BA 9, bordering BA 8 in standardized space.
The findings by Herwig et al (2003), therefore, were consistent with previous studies (Gerloff,
180
Corwell et al. 1997; Rossi, Cappa et al. 2001) that identified electrode F3 as the closest electrode
to target the left DLPFC. Although using the F3 electrode (‗10-2 method‘) to represent the left
DLPFC may be an improvement from the conventional ‗5-cm method‘, it is still limited in
several ways. First, as the DLPFC is a functional area, a comparison between an individual T1-
weighted MRI and the Talairach atlas may not adequately identify this area. Second, although
the ‗10-20 method‘ takes into account head size, it does not take into account head shape, an
important consideration in studies using rTMS therapeutically. Third, this method also does not
take into consideration the morphology of the underlying cortex so that simply placing TMS coil
over F3 to target the left DLPFC is less than ideal. By contrast, localizing the DLPFC based on
functional standardized coordinates (e.g. Talairach or MNI coordinates) and then determining its
relative location on the scalp for TMS coil placement may enhance the likelihood that the desired
cortical region is activated. This study, therefore, was designed to evaluate the accuracy of
previous methods to localize the DLPFC compared to our novel neuronavigational method,
which first estimates the cortical position based on a standard template and then determines the
most appropriate position for the TMS coil on the scalp in healthy subjects.
7.3 Methods and Materials
Fifteen right-handed healthy volunteers were tested (mean age=35.1 years, SD=7.77
years, range=24-49 years; 10 men and 5 woman). Subjects gave their written informed consent
and the protocol was approved (Centre for Addiction and Mental Health in accordance with the
declaration of Helsinki). Exclusion criteria included psychiatric or medical illness, a history of
drug or alcohol abuse.
181
7.3.1 Procedure
7.3.1.1 Magnetic Resonance Image (MRI)
A T-1 weighted MRI (Acquisition Type=3D, TR=8.892 ms, TE=1.792 ms, Inversion
Time=300, Number of Averages=1, Slice Thickness=1.5, FOV=20 cm, Matrix=256x256) was
acquired for all subjects with seven fiducial markers in place for future co-registration. The
images were converted to isotropic voxels of side 0.86 mm and the position of the anterior
commissure (AC) was identified. The images were then stored in 16 bits with values ranging
from 0 to 3000. A threshold value of 250 defined the iso-surface representing the scalp.
7.3.1.2 Functional Coordinates for the left DLPFC
The results from recent meta-analyses on functional neuroimaging studies testing
working memory in patients with depression and schizophrenia (Glahn, Ragland et al. 2005;
Mendrek, Kiehl et al. 2005; Tan, Choo et al. 2005; Fitzgerald, Oxley et al. 2006) were used to
identify the functional region of the left DLPFC. These studies, however, localized the left
DLPFC to an area within the anterior portion of the medial prefrontal cortex at the juncture of
BA 9/46, and did not specify a single coordinate position. Although a study conducted earlier by
our group examined rTMS applied to DLPFC position using (x, y, z) = -45,45,35 in Talairach
coordinates ((x, y, z) = -46, 45, 38 MNI coordinates) resulted in greater reduction in depressive
scores in MDD patients, this does not preclude that other voxel positions within this juncture
may more optimally improve the antidepressant effects of rTMS in MDD patients. As such, we
selected the coordinate (x, y, z) = -50, 30, 36 after converting these coordinates to MNI space
that also corresponded to the juncture of BAs 9 and 46. Our DLPFC coordinate (x, y, z) = -50,
30, 36 mm resulted in a position that was a little more posterior and lateral to those examined by
Fitzgerald et al. (2009).
182
7.3.1.3 Optimal TMS Coil Placement for Targeting the Left DLPFC
Importantly, the neuronavigational method used to position the TMS coil on the scalp in
this study, can be applied to any cortical coordinates in the brain. It involved two steps: we first
found the position of the left DLPFC on the cortex of each subject (DLPFCC) based on the
normalization of a template, followed by the application of a marching cubes algorithm to find
the optimal position for the coil on the scalp (DLPFCS; Figure 1). These procedures were done
for all 15 subjects included in this study.
7.3.1.3.1 Estimating the left DLPFC on the cortex (DLPFCC)
Firstly the DLPFC (x, y, z=-50, 30, 36; MNI brain) was identified on the brain template
MNI/ICBM 152 (Evans 1993; Mazziotta, Toga et al. 2001) by meta-analyses on neuroimaging
studies examining working memory in patients with MDD and schizophrenia (Glahn, Ragland et
al. 2005; Mendrek, Kiehl et al. 2005; Tan, Choo et al. 2005; Fitzgerald, Oxley et al. 2006).
These coordinates were also shown to result in superior therapeutic efficacy in MDD when
targeted through rTMS compared to the ‗5-cm method‘ (Fitzgerald, Maller et al. 2009). For each
subject, the best non-linear transform to map the standard brain to each individual brain was
estimated using the subroutine spm_normalize from the SPM2 (www.fil.ion.ucl.ac.uk/spm/;
(Ashburner, Neelin et al. 1997; Ashburner and Friston 1999). Next, the coordinates for the
DLPFCC in individual brains were estimated by applying the resulting transformation to the
coordinates of the DLPFCC in the template (Figure 1; Step 1A).
183
Figure 2. A non-linear transformation to convert the coordinates from a standard space to an
individual space was first estimated (Step 1A). This estimated transform was then applied to the
coordinates for the DLPFC in the standard brain (or template) to determine the position of the
DLPFC in the individual brain (Step 1B). Next, a triangular mesh wrapping the iso-surface
representing the scalp was created with the marching cube algorithm (see Methods). The
optimal position for the coil (mDLPFCS) on the scalp was then defined as the vertex of the
triangular mesh with a normal that passed through the DLPFC on the cortex (DLPFCC; Step 2).
184
7.3.1.3.2 Estimating the position of the left DLPFC on the scalp (DLPFCS)
The DLPFCS on the scalp was defined as the position in which the tangential plane to the
scalp had a perpendicular (called normal) that passed through the DLPFCC. The position on the
surface of the scalp was estimated using a classic algorithm for rendering called marching cubes
(Lorensen 1987) followed by a selection process. To use the marching cubes algorithm, a
threshold value of 250 (the value-tone color at the border of the scalp) was chosen to define an
isosurface on the MRI representing the scalp. The marching cubes algorithm
(http://local.wasp.uwa.edu.au/~pbourke/geometry/polygonise/) generates a triangular mesh (a set
of 3-3D coordinates) rendering this 3D surface on the image. Note that in actuality this threshold
value defines 3 surfaces in a T1 contrast: the exterior surface of the scalp, the interior surface of
the skull and the surface of the cortex. In addition, to create the triangular mesh, this version of
the marching cube algorithm determined a normal to the iso-surface at each vertex of the mesh
averaging the normals to the faces of all the triangles that shared this vertex.
The selection process to determine the location of the DLPFCC on the scalp consisted of 7
sequential steps:
1) The coordinates of each vertex was associated with the closest voxel and only the voxels
with a normal that intersected a sphere of 3 mm around the DLPFCC at a distance less
than 3 cm were considered.
2) For each voxel ( i ) the average distance ( id ) to the other voxels were calculated. If
mmdi 23 , the voxel ( i ) was filtered out. This filtering process was necessary to
remove voxels that were generated by the marching cubes algorithm that were further
from the main cluster and were likely to be outliers.
185
3) A second filter was then used to remove voxels that were within 1.5 cm of the DLPFCC,
as these voxels were most likely to be on the cortex rather than the scalp.
4) Next, an average normal from all of the ensuing voxels (Figure 2A) was calculated, and
those voxels with a normal that formed an angle greater than 25° with respect to the
average were removed.
5) An average position and normal was then calculated from the remaining vertices.
6) A 3D image of a 6 cm line that represented a normal to the scalp (estimated in step 5)
which also passed through the middle point of the vertices‘ mean position was then
superimposed on the MRI image. As quality control this line should have passed through
the DLPFCC.
7) The position of the coil on the scalp (DLPFCS) was then determined by manually by
selecting one point on this 3D line, which fell on the scalp. A spherical mark with a
diameter of 2.5 mm (i.e., small than a fiducial mark) was then superimposed on the MRI
image for later co-registration of the MRI image (Figure 2B). The position for the
DLPFCS predicted through our method will be called ‗method’ DLPFCS (mDLPFCS),
while this position is referred to as the ‗experimental’ DLPFCS (eDLPFCS) once it is co-
registered to the subjects‘ MRI, representing its true position on the scalp.
186
Figure 2. A) In this coronal slice of a MRI image, the orange spot represents the DLPFC on the
cortex (DLPFCC) and the green squares are voxels on the iso-surface with a normal that passes
through the DLPFCC in a distance less than 3 cm. Voxels (green squares) on surfaces other than
the scalp (i.e., the internal side of the skull, and the interface gray matter-CSF) were filtered out
if they had a normal greater than 25° from the average normal. The mDLPFCS was defined as
the intersection of the scalp with a line with the orientation of the average normal and passes
through their average position. B) The position of mDLPFCS was superimposed on each
individual's MRI for later co-registration. Once mDLPFCS is co-registered, this position is
referred to as the eDLPFCS.
187
7.3.1.4 MRI co-registration
Prior to MRI co-registration, a 64 channel EEG cap (STIM2, Neuroscan, U.S.A.) was
placed on the subjects‘ head to mark the positions of electrodes AF3, F3, F5, FC3, and FC5 (‗10-
20 method‘) with an ink-filled syringe. The EEG cap was then removed for identification of the
‗5-cm method‘.
In accordance to the ‗5-cm method‘ to localize the DLPFC, single monophasic TMS
pulses (Magstim Company Ltd., UK) were administered to the left APB of the motor cortex,
while resulting electromyography activity was collected using commercially available software,
Signal (Cambridge Electronics Design, UK). A felt pen was then used to mark the optimal
position under the centre of the coil for eliciting motor evoked potentials from this muscle (APB)
and the location of the DLPFC was measured 5 cm anterior from this position.
Neuronavagational techniques (MINIBIRD system; Ascension Technologies) combined
with MRIcro/reg software were first used to co-register the 7 fiducial markers to subjects‘ MRI
with the position of the mDLPFCS superimposed on the image (represented as a sphere in Figure
2B). The position of the DLPFCS was then marked on the subjects‘ scalp and its experimental
coordinates recorded, and referred to as eDLPFCS. Next, the positions marked on the scalp for
APB, ‗5-cm method‘, AF3, F3, F5, FC3, and FC5 electrodes (‗10-20 method‘) were registered to
the image and their coordinates recorded. These coordinate positions were first translated to the
MNI/ICBM using an affine transformation from the individual brain to the template (i.e., the
inverse of the affine transformation that was first used in the non-linear transformation to
estimate DLPFCC). Finally, the positions APB, ‗5-cm method‘, and AF3, F3, F5, FC3, FC5
electrodes (‗10-20 method‘) relative to the eDLPFCS were also measured to provide data in
individual space.
188
7.3.1.5 Software implementation
The algorithms were programmed in C++ and integrated in a friendly graphic user
interface. SPM ran under MATLAB (The MathWorks, Inc. Natick, MA, USA), called from the
C++ via the API interface. In total our method takes approximately 5 minutes with most of the
time allotted to selecting the voxel location of the mDLPFCS (Step 7). The automated steps are
practically instantaneous with any current hardware. Software is available by request via e-mail
7.3.2 Data Analysis
To evaluate the inter-subject variability found between all of the methods (i.e., ‗5-cm
method‘, ‗10-20 method‘, and our method), a jackknife approach was used for each ellipsoid
volume representing the dispersion (N=15) of each landmark:
1) The mean position and the standard deviation of each dimension ( j
z
j
y
j
x ,, ) for each
method (condition) in each sub sample ( j ), by leaving one subject ( j ) out of the
jackknife, where j=1-15 was calculated. The standard deviations were then used to
estimate the dispersion as the volume of an ellipsoid:
j
z
j
y
j
x
j
ellipsoidVol 3
4
2) The difference between the dispersion of methods within each sub sample was then
calculated as the difference of ellipsoid volume.
3) Finally, Wilcoxon non-parametric signed-rank tests were used to determine whether
the volumes derived under each method (i.e. the 3-dimensional amount of variability
in estimates) differed. A non-parametric approach seemed more appropriate than a
189
series of paired t-tests, since the assumptions underlying parametric tests most likely
would not have been met due to our sample size limitation. In addition, it may not
have been appropriate to assume that the distributional properties of the composite
standard deviation volume measurements were normally distributed.
7.4 RESULTS
7.4.1 Individual space
The positions of APB, the ‗5-cm method‘, and electrodes AF3, F3, F5, FC3, and FC5
(‗10-20 method‘) relative to the eDLPFCS were measured with a measuring tape and are shown
in Table 1. These measures are not normalized by head size, and thus, provide an evaluation of
both the ‗5-cm method‘ and the ‗10-20 method‘ to localize the left DLPFC in individual space.
The average distance from APB to the eDLPFCS was approximately 5.3 cm, ranging from 3.0-
7.3 cm in 15 subjects. Although 5.3 cm is close to 5cm, this position was not on the same
sagittal plane as the position given for the ‗5-cm method‘, and the average distance between
these two landmarks was 2.6 cm. In evaluating the ‗10-20 method‘ to localize the left DLPFC,
we found that F5 was the closest electrode to eDLPFCS with a mean difference of 1.9 cm,
followed by electrodes FC5 (2.3 cm) and F3 (2.8), respectively (Table 1). Table 2, compares the
voxel coordinates predicted by our method (i.e., the; Figure 2B) mDLPFCS with the position
obtained with the co-registration probe during the experiment, referred to as eDLPFCS. Further,
the difference between these two positions was 12.6 mm with the x-direction (left-right) driving
this difference. More explicitly, the experimental coordinates were found to be more lateral (i.e.,
more exterior) than the centre of the fake fiducial marking the position of mDLPFCS (see Figure
3, same result in standardized space). This discrepancy may reflect some systematic bias of the
method (i.e., manual selection of the voxel representing mDLPFCS) that could be related to the
190
intrinsic spatial resolution pattern of the MINIBIRD system, accuracy of the co-registration
procedures or the need to take into account the physical characteristics of the scalp (i.e., amount
of hair on the scalp, size of the probe). As such, the difference between these two locations (i.e.,
mDLPFCS and eDLPFCS) will be statistically tested in standardized space to account for
differences arising from head size and MRI orientation.
7.4.2 Standardized space
In standardized space, the closest position to eDLPFCS was found with the ‗5-cm
method‘ (Table 3). However, the average distance of eDLPFCS to APB was approximately 7 cm
and not along a parasaggital line as in the ‗5-cm method‘. Figure 4 illustrates that DLPFCS
corresponds best to a position more lateral, inferior and posterior than F3, and superior to F5. In
evaluating the inter-subject variability in ‗5-cm method‘ and the ‗10-20 method‘ compared to our
method, Wilcoxon non-parametric signed-rank tests were performed (SAS System v.9.1.3; SAS
Institute, NC, USA). Our method yielded the least amount of variation compared to the ‗5-cm
method‘ (p<0.0001) and compared to the ‗10-20 method‘ (p<0.0001) for both electrodes F3 and
F5, respectively. In addition, less inter-subject variability was found with ‗10-20 method‘
compared to the ‗5-cm method‘ (p=0.0002, electrode F3; p≈0.000, electrode F5, respectively).
Finally, with all three methods, a greater standard deviation was observed in the anterior-
posterior direction that may be reflective of the pattern of spatial resolution of the MINIBIRD,
however, our method still yielded the least inter-subject variability for eDLPFCS compared
positions generated by the other methods.
Figure 3 compares the coordinates predicted by our method mDLPFCS with the position
obtained with the co-registration probe during the experiment, referred to as eDLPFCS in
standardized space. A square T-test (SAS System v.9.1.3; SAS Institute, NC, USA) was
191
performed between the eDLPFCS and mDLPFCS and found that the locations differed
significantly (T-square=46.44, df=3,12, p=0.0004). Furthermore, a series of paired t-tests found
that location of the mDLPFCS and eDLPFCS differed significantly in the x-dimension
(p=0.0004), moderately in the y-dimension (p=0.0651), while there was no difference in the z-
dimension (p=0.2947).
192
Figure 3. Although the position predicted by our method (mDLPFCS; green) is placed on the
subjects‘ scalp, the actual position measured with the probe when co-registered (eDLPFCS;
orange) is biased in the radial direction. These positions are shown for all subjects in MNI
space.
193
Table 1: Actual distance (cm) from the experimental location of left DLPFC (eDLPFCS) to EEG
electrodes, APB, and the ‗5-cm method‘ on the scalp in all 15 healthy subjects in individual
space.
Table 2: A comparison showing the voxel coordinates predicted by our method (mDLPFCS)
with the voxel coordinates obtained during the experiment with the co-registration probe
(eDLPFCS) in individual space.
194
Table 3: Positions in standard MNI space (mm) for each subject predicted by the method
(mDLPFCS) compared to positions measured during the experiment (eDLPFCS).
195
196
Figure 4. The dispersion of APB, ‗5-cm method‘, eDLPFCS and electrode positions are
represented by spheroids in standardized space. Spheroids are centered at the mean position for
each measurement with a radius in each direction (i.e. X, Y, and Z) equal to 1 standard deviation.
Please note that these positions were measured on the scalp so they are ―flying‖ over the gray
matter.
197
7.5 Discussion
Here we presented a neuronavigational method to determine the most appropriate
position to place the TMS coil on the scalp in order to target left DLPFC, and compared this to
the ‗5-cm method‘ and ‗10-20 method‘. We found that our neuronavigational method was the
least susceptible to inter-subject variability followed by the ‗10-20 method‘ and lastly the ‗5-cm
method‘ in standardized space. Furthermore, we found that our neuronavigational method
yielded a DLPFC location more anterior and lateral than the ‗5-cm method‘ and between
electrodes F3 and F5 of the ‗10-20 method‘. However, if MRI acquisition is not feasible, we
recommend using the ‗10-20 method‘ by positioning the TMS coil over the F5 electrode as we
found that the ‗10-20 method‘ was less susceptible to inter-subject variability compared to the
‗5-cm method‘.
Consistent with Herwig and colleagues (Herwig, Padberg et al. 2001), inter-subject
variability was found with the ‗5-cm method‘, which was greater than the variability found with
our method (eDLPFCS position) and the ‗10-20 method‘ (Table 3; Figure 4). Also consistent
with Herwig and colleagues (Herwig, Padberg et al. 2001), we found that the ‗5-cm method‘
generated positions for the DLPFC that were posterior of our position for the DLPFC based on
meta-analyses on neuroimaging DLPFC activation during working memory in patients with
depression and schizophrenia (Glahn, Ragland et al. 2005; Mendrek, Kiehl et al. 2005; Tan,
Choo et al. 2005; Fitzgerald, Oxley et al. 2006). Thus, the recommendations in this study are
based on patients with depression and schizophrenia; however, the advantage of our
neuronavigational method is that it can be employed to target any brain region.
The ‗10-20 method‘ was found to be less susceptible to inter-subject variability compared
to the ‗5-cm method‘ and yielded positions closer to the DLPFC position obtained by our
198
method. Furthermore, when MRI coregistration is unavailable, the ‗10-20 method‘ should used
to direct the TMS coil over the F5 electrode (Figure 4; individual space). This finding, therefore,
suggests that the TMS coil should be centered on electrode F5 and by extension F6 (‗10-20
method‘) to target the left and right DLPFC, respectively.
Our group has previously shown that targeting the DLPFC using neuronavigational
techniques based using functional coordinates (x, y, z) = -45, 45, 35 (Talairach coordinates; (x, y,
z) =-46, 45, 38; MNI coordinates) for this area results in a significant reduction in scores on the
Montgomery-Asberg Depression Rating Scale compared to those patients whose DLPFC was
determined through the standard or ‗5-cm method‘ (Fitzgerald, Hoy et al. 2009). The functional
coordinates investigated in this study were guided by a previous meta-analysis on neuroimaging
studies which examined working memory in patients with major depressive disorder (MDD;
(Fitzgerald, Oxley et al. 2006)). The results of the meta-analysis found the DLPFC to be located
within the juncture of BAs 9 and 46. Although, using this coordinate position indeed resulted in
greater treatment efficacy, it does not preclude that other voxel coordinates within this juncture
may more optimally improve the anti-depressant effects of rTMS in MDD patients. In this
regard, we selected our voxel coordinate for the DLPFC by first converting the coordinates used
by Fitzgerald et al. (2009) to MNI coordinates that also were associated within the juncture of
BAs 9 and 46. Our DLPFC coordinate resulted in a position that was a little more posterior and
lateral to those examined by Fitzgerald et al. (2009). In line with our position for the DLPFC, a
recent study (Herbsman, Avery et al. 2009) who examined the relationship between the
variability of the DLPFC location determined by the 5 cm rule (i.e., ‗5-cm method‘) with scores
on the Hamilton Depression Rating Scale in MDD patients. Herbsman et al. observed greater
clinical response to rTMS in individuals with DLPFC positions that were more lateral and
anterior compared to individuals with more medial and posterior locations. The mean position of
199
the DLPFC was (x, y, z) =-46, 25, 44 (Talairach coordinates; (x, y, z) = -47, 24, 48; MNI
coordinates) in those subset of subjects with more lateral and anterior positions. The DLPFC
position examined in the current study (x, y, z) = -50, 30, 36; MNI coordinates) and the one in
Herbsman et al. (x, y, z) = -47, 24, 48; MNI coordinates), thus, are more lateral and posterior to
the coordinate previously investigated by our group (Fitzgerald, Hoy et al. 2009), and suggests
that a more lateral position within the junction of BAs 9 and 46 may more optimally improve
treatment efficacy in MDD.
The results of this study are limited in several important ways. First, our proposed
method for determining the position of DLPFCS assumes that the centre of the coil makes
contact with the scalp in a single point therefore only one plane that was tangential to the scalp at
this point were considered (Figure 3). However, hair and skin may have produced a different
contact surface that may have caused some range of possible orientations for the TMS coil rather
than a single tangential plane. This is consistent with the difference observed between the
mDLPFCS and eDLPFCS locations in the x-dimension, which may have arisen from step 7
during the mDLPFCS voxel selection, which is manually selected. In this step the scalp DLPFC
location in the MRI was delineated as an arbitrary value of color intensity. A change in this
threshold could move the scalp to a maximum in a voxel in either direction (1 voxel=0.86 mm).
The true difference observed in the x-dimension between the two locations was equal to 10.4 mm
in standardized space of which the size of the coregistration probe (8 mm), in addition to hair and
skin may have contributed to this difference. The difference found between mDLPFC and
eDLPFC positions, therefore, reflects a distance along the normal to the tangential plane that the
TMS coil is positioned. Most importantly, regardless of where which voxel for the mDLPFC we
select the along this line, the effect of the TMS on the DLPFC would not change and therefore is
clinically insignificant. We have determined, however, that by selecting a voxel position that
200
lies above the scalp rather than one that intersects the scalp, the difference arising in the x-
dimension will be minimized. For example, in Figure 2B we can superimpose a line
representing the normal to the tangential plane on the scalp instead of a sphere. With this slight
modification of our voxel selection process, any differences between the mDLPFC and eDLPFC
positions would be owing to the variability in the MINIBIRD system itself. Second, since the
non-linear transformation (SPM) used in this study to convert the standard template to the
individual brain (Figure 1; Step 1) were designed to match the full brain rather than just the
cortex, other transforms that are based on the trajectory of deep sulci in the cortex (i.e., surface-
based techniques for warping three-dimensional images of the brain (Thompson and Toga 1996))
may increase the accuracy of the method proposed in this study. Future studies examining the
accuracy in using non-linear transformations for determining the position of Talairach
coordinates to individual cortex should be investigated. Furthermore, future studies should also
aim to improve the algorithms used for rendering the surface of the scalp. Although the
marching cubes algorithm (based on an arbitrary threshold value) seemed to work very well, this
tends to render other iso-surfaces (i.e., internal side of the skill and on the cortex) rather than just
the iso-surface of the subject‘s scalp. Removing these extra iso-surfaces would decrease the
filtering processes and possible bias involved when the voxel for the mDLPFCS, is selected and
therefore improve upon our method.
Although previous studies have also employed the marching cubes algorithm to
determine scalp positions to target other brain regions (Andoh, Riviere et al. 2009), this is the
first demonstration of neuronavigational methods which took into consideration the functional
coordinates of the brain region of interest and the orientation of the TMS coil in order to
optimally target the DLPFC. We used a novel neuronavigational method, which accurately
targeted the DLPFC with less inter-subject variability compared to both the ‗5-cm method‘ and
201
‗10-20 method‘. Our method improves upon these previous methods to localize the DLPFC in
two respects. First, although neuroimaging studies report that the location of the DLPFC at the
juncture of BAs 9 and 46, our group previously demonstrated that using functional coordinates
more accurately determines the most appropriate position which to direct the TMS coil in order
to optimize rTMS treatment efficacy in MDD (Fitzgerald, Hoy et al. 2009). We, therefore,
account for the fact that this is a functional brain region that is not considered in either the ‗5-cm
method‘ or the ‗10-20 method‘. Second, a marching cubes algorithm determined the optimal
location for TMS coil placement on the skull from the cortical position, thereby accounting for
cortex morphology, skull shape, and the orientation of the TMS coil. We, therefore, have
minimized inter-subject variability inherent in these methods that do not consider characteristics
of head and/or cortex or TMS coil orientation. Our neuronavigational method, therefore,
demonstrates a significant advancement to accurately target the DLPFC in rTMS treatment
studies.
The method proposed in this study is important for several reasons. First, previous
studies have shown that targeted rTMS to the DLPFC results in greater clinical response in
patients with MDD compared to non-targeting techniques (Fitzgerald, Hoy et al. 2009). Second,
F5 from the ‗10-20‘ system was found to be the closest electrode to target the DLPFC and may
be used in clinical practice without an acquisition of a MRI. Finally, our method is straight-
forward, requires approximately 5 minutes to perform, and can be applied to any brain region.
202
Copyright Acknowledgements (if any)