www.sciencedirect.com

c o r t e x 4 9 ( 2 0 1 3 ) 6 9 1e7 0 1

Available online at

Journal homepage: www.elsevier.com/locate/cortex

Research report

Oculomotor deficits affect neuropsychological performancein oculomotor apraxia type 2

Silvia Clausi a,b,1, Maria De Luca c,1, Francesca R. Chiricozzi a,b, Anna M. Tedesco a,b,Carlo Casali d, Marco Molinari b and Maria G. Leggio a,b,*aDepartment of Psychology, University of Rome ‘Sapienza’, Rome, ItalybAtaxia Laboratory, Santa Lucia Foundation, IRCCS, Rome, ItalycNeuropsychology Unit, Santa Lucia Foundation, IRCCS, Rome, ItalydDepartment of Medical and Surgical Science and Biotechnologies, University of Rome ‘Sapienza’ e Polo Pontino I.C.O.T., Latina, Italy

a r t i c l e i n f o

Article history:

Received 21 July 2011

Reviewed 26 September 2011

Revised 22 November 2011

Accepted 21 February 2012

Action editor Georg Goldenberg

Published online 6 March 2012

Keywords:

AOA2

Cognition

Eye movements

Reading

Saccadic intrusions

* Corresponding author. Department of PsycE-mail address: maria.leggio@uniroma1.i

1 S.C. and M.D. contributed equally to this0010-9452/$ e see front matter ª 2012 Elsevdoi:10.1016/j.cortex.2012.02.007

a b s t r a c t

Introduction: Ataxia with oculomotor apraxia type 2 is a rare and early-disabling neurode-

generative disease, part of a subgroup of autosomal recessive cerebellar ataxia, in which

oculomotor symptoms (e.g., increased saccade latency and hypometria) and executive

function deficits have been described.

The aim of this study was to evaluate the impact of oculomotor symptoms on cognitive

performance and, in particular, over reading in 2 Italian siblings affected by ataxia with

oculomotor apraxia type 2.

Methods: The neuropsychological profiles and the oculomotor patterns during nonverbal

and verbal tasks were recorded and analyzed.

Results: Saccadic intrusions and/or nystagmus were observed in all eye movement tasks.

The neuropsychological profiles were substantially preserved, with only subtle deficits that

affected visuomotor integration and attention. Reading ability decreased and became

impaired. The reading scan was disturbed by saccadic intrusions and/or nystagmus.

However, an ad hoc reading task demonstrated that deficits appeared only when the items

that were displayed enhanced oculomotor requests. The preservation of lexical-semantic

processes confirmed that the reading disability was caused by oculomotor deficits, not

cognitive problems.

Conclusion: Present findings indicate that in patients who are affected by ataxia with

oculomotor apraxia type 2, performance on neuropsychological tests, especially those that

require rapid performance and eye or handeeye control, must be analyzed with respect to

oculomotor components.

ª 2012 Elsevier Ltd. All rights reserved.

hology, University of Rome ‘La Sapienza’, Via dei Marsi 78, 00185 Roma, Italy.t (M.G. Leggio).work and should be considered co-first authors.ier Ltd. All rights reserved.

c o r t e x 4 9 ( 2 0 1 3 ) 6 9 1e7 0 1692

1. IntroductionTable 1 e Clinical findings.

Clinical findings Patient 1 Patient 2

Sex/age at onset y/age at

examination y

F/20/38 M/15/40

Sign at onset Difficulties

in writing

Gait

in-coordination

Visual acuity 11/10 11/10

Dystonic movements D

Choreic movements in

superior limbs

D

Dysphagia D D

Diplopia D

Menstrual disorders D

Areflexia D D

Distal motor deficit D D

Hand and Foot deformity D

Impaired position/

vibration sense

D D

Impaired superficial senses D

Distal sock hypesthesia D D

Volitional dyskinesias D D

Gait ataxia D D

Dysmetria D

Hypotonia D D

Dysarthria D D

Cerebellar atrophy D D

Ocular motor apraxia D D

Urge incontinence D

Axonal neuropathy D D

Serum a-fetoprotein 50.71 ng/ml 39.93 ng/ml

Ataxia with oculomotor apraxia type 2 (AOA2) is a rare and

early-disabling neurodegenerative disease that, with ataxia

with oculomotor apraxia type 1 (AOA1), belongs to a subgroup

of oculomotor apraxia-associated autosomal recessive cere-

bellar ataxia (ARCA) (Moreira et al., 2001, 2004).

The onset of AOA2 occurs between age 10 and 22 years

(Criscuolo et al., 2006; Le Ber et al., 2004; Tazir et al., 2009).

Elevated serum a-fetoprotein and creatine kinase (CK)

concentrations and cerebellar atrophy have been reported

(Criscuolo et al., 2006; Le Ber et al., 2004). AOA2 is character-

ized by optional oculomotor apraxia (saccade of elevated

latency due to a failure to initiate the saccade present in about

50% of subjects), peripheral neuropathy, and extrapyramidal

signs, including choreiform movements, dystonia, and

tremor. Recently, in line with the importance of cerebellum in

cognition (Leggio et al., 2011), cognitive impairments have

been also reported (Le Ber et al., 2004). However, most neu-

ropsychological tests require unimpaired visual scanning

abilities that might be affected by AOA2-induced oculomotor

deficits.

In this study, we analyzed the neuropsychological profiles

of 2 Italian siblings who were affected by AOA2 and their

oculomotor patterns during nonverbal and verbal tasks to

determine the influence of impairments in visual scanning on

cognitive performance.

concentration (normal

value <10 ng/ml)

ICARS subscores

Posture and gait disturbances 31 30

Kinetic functions 21.5 24

Speech disorders 5 5

Oculomotor disorders 5 5

Global score 62.5 64

ICARS ¼ International Cooperative Ataxia Rating Scale. The

symbol þ indicates presence of feature.

2. Subjects

We studied 2 patients from a family in central Italy, patient 1

and patient 2, who were affected by AOA2, harboring a large

homozygous deletion that encompassed 8 exons in senataxin

(SETX) gene.

Patient 1 was a 38-year-old woman, and patient 2 was a 40-

year-old man. Both patients were right-handed and had

13 years of education. The pedigree of the family and its

molecular data have been reported by Criscuolo et al. (2006)

(Patients 3 and 4 in Family 3). The procedures were approved

by the Santa Lucia Foundation Ethical Committee; written

consent was obtained from the participants according to the

Declaration of Helsinki.

The clinical history of both patients was obtained, and

a neurological examination (including a motor scale)

(Trouillas et al., 1997), routine laboratory tests, peripheral

nerve conduction tests, and brain magnetic resonance

imaging (MRI- including Spin-Echo, T1- and T2-weighted

images) were performed. The detailed clinical findings are

summarized in Table 1.

Patient 1 developed symptoms at age 20 years, manifested

as difficulties in writing. At age 36 years, she required

a wheelchair and presented with dystonic movements,

dysphagia, diplopia, and menstrual disorders. Her neurolog-

ical examination revealed generalized areflexia, moderate

distal motor deficits, and deformities of the hands and feet

(abnormally high plantar arch and hyperextension of the toes:

“pes cavus”). Positional and vibrational senses were impaired

in the inferior limbs, and superficial senses were reduced,

accompanied by distal sock hypesthesia. Volitional dyskine-

sias of the right hand were evoked by voluntary movements.

Cerebellar ataxia, hypotonia, and dysarthria were evident. A

brain MRI showed diffuse cerebellar atrophy, more evident in

the vermis (Fig. 1a).

Patient 2 developed symptoms at age 15 years, as evi-

denced by an uncoordinated gait. He had been using

a wheelchair since age 35 years. The neurological examina-

tion revealed choreic movements in the superior limbs, voli-

tional dyskinesias of the fingers, generalized areflexia, distal

motor deficit, impaired vibrational sense in the inferior limbs,

and distal sock hypesthesia. He had cerebellar ataxia, a wide-

based stance, generalized hypotonia, dysmetria, and dysar-

thria. Episodic dysphagia and urge incontinence were also

present. His brain MRI showed marked cerebellar atrophy of

the vermis (Fig. 1b).

In the clinical examination of both patients, eye move-

ments were abnormal with inconstant oculomotor apraxia,

horizontal ocular oscillation, and saccadic pursuit. Moreover,

Fig. 1 e T1-weighted sagittal brain MRI and T2-weighted coronal MRI sequences of patient 1 (a) and patient 2 (b). Both

subjects present with diffuse cerebellar atrophy which predominates in the vermis. This pattern is more evident in patient 1

than in patient 2.

c o r t e x 4 9 ( 2 0 1 3 ) 6 9 1e7 0 1 693

they had axonal sensorimotor neuropathy and high serum

a-fetoprotein concentrations. At the time of testing both

patients were free of any medication.

3. Methods

3.1. Eye movement recordings

3.1.1. Apparatus and general procedureEyemovements were recorded at a sample rate of 500 Hz from

the dominant eye (Porac and Coren, 1981) in binocular vision

using an infrared eye tracker (for details, see De Luca et al.,

1999). The participant sat in front of a 1500 computer screen

(60-cm eye-screen distance), with the head fixed. A calibration

was run before each experimental trial, acquiring gaze posi-

tion only during steady fixation and excluding intrusive

movements.

A simple fixation task was used to evaluate the ability to

maintain gaze over a fixed target (a white dot inscribing a red

cross, subtending .4� of visual angle) that was shown on

a black background at the center of the screen. The task

required the patient to look steadily at the center of the cross

for 10 sec; 6 trials were run.

A saccade task was used to assess the latency of eye

movements in response to amoving target. The target (a black

dot, subtending .2� of visual angle), displayed on a white

background, appeared along the horizontal meridian in 5

consecutive positions 4.0� to each other, according to

a synchronous paradigm (i.e., no gap) in a left-to-right

sequence and vice versa, repeated twice (for details, see De

Luca et al., 1999). The task was to saccade to the dot as soon

as it appeared. Three trials were run.

The reading task comprised 64 high-frequencywords and 64

nonwords. For both stimuli, 2 blocks with short items (4e5

letters) and 2 with long items (8e10 letters) were presented

(for details, see De Luca et al., 2002). Each block contained 16

c o r t e x 4 9 ( 2 0 1 3 ) 6 9 1e7 0 1694

items, displayed in 4 rows at the center of the screen. The task

required to read the text silently without a time limit. After

each trial, the experimenter read 4 items, and the participant

indicated the 2 that were part of the list.

3.1.2. Data analysisDuring the offline analysis, only fixations that matched the

experimental calibration points were used to transform raw

eye movements into gaze position data. Blinks and artifacts

were rejected. Quantitative analyses were performed for the

horizontal component of eye movement. Gaze position, fixa-

tions, and saccades were detected manually by visual

inspection of the traces for all eye movement tasks. The

patients’ data were compared with those of their healthy 35-

year-old sister, who had a normal genotype and normal clin-

ical assessment and no history of neurological disease.

3.1.2.1. FIXATION TASK. The frequency (number/minute),

amplitude (degrees), velocity (degrees/second), and duration

(milliseconds) of horizontal saccadic intrusions (SI) were

measured. T-tests for independent samples were applied to

compare amplitude, velocity, and duration measures with

control data and between patients. Chi-square test was used

to analyze the frequency of SI.

3.1.2.2. SACCADE TASK. Saccade latency was measured,

excluding latencies of anticipatory saccades from the anal-

ysis. Due to the presence of SI, latency data were validated

only for saccades that did not belong to trains of SI. T-tests for

independent samples were used for comparisons with control

data. For each participant, t-tests for repeated measures were

performed separately to examine latency differences between

rightward and leftward directions of the task.

3.1.2.3. READING TASK. The mean number of fixations per item

and mean fixation duration (separately for words

and nonwords, short and long) were calculated, netted of SI.

T-tests for independent samples were used for comparison

with control data.

3.2. Neuropsychological examination

3.2.1. Neuropsychological assessmentAn extensive neuropsychological battery was used to evaluate

the following cognitive domains: visuospatial abilities,

language abilities, executive functions, verbal and visuospatial

memory, visual-motor integration (VMI), and attention. Intel-

ligence was also measured. The tests (Beery, 1997; Borkowsky

et al., 1967; Carlesimo et al., 1996; Corsi, 1972; De Renzi and

Faglioni, 1978; Gainotti et al., 1977, 2001; Gauthier et al., 1989;

Miceli et al., 1994; Orsini and Laicardi, 2001; Raven, 1947; Rey,

1958; Shallice, 1982; Villa et al., 1990; Wechsler, 1997;

Weintraub and Mesulam, 1985; Zimmermann and Fimm,

1995) are detailed in Table 2.

3.2.2. Additional reading and vocal testsTo better assess reading abilities, we used a test that was based

ondifferences in text arrangement: thenew-Developedwordsand

nonwords Reading task (nDR task). In the nDR task, stimuli (words

or nonwords) were selected from the battery for evaluating

dyslexia and dysorthography (subtests 4 and 5) (Sartori et al.,

1995). The width of each stimuli ranged between .8 and 2.3�.The stimuli were presented on a 1700 computer screen (black

letters in courier on a white background), arranged in columns

(vertical display) or as plain text (horizontal display). In the

vertical display, 4 lists (2 containing 23 words and 2 containing

23 nonwords) were presented separately, alternating between

8 cm left or right of the screen center. In the horizontal display,

55 words and 55 nonwords were presented at the monitor

center in2 sessions, eachcontaining7 rows, subtendinganarea

of 16� 13�. Reading time (sec/item) and percentage of errors

were computed for each subtest.

The simple vocal reaction time task was used to assess visual

alertness under conditions (uncued and cued) that required

only a vocal response (for details see Spinelli et al., 2002).

Median vocal reaction times (RTs) were measured.

The word length effect was evaluated by the vocal reaction

time to 4e7-letter words task (modified from Zoccolotti et al.,

2006). Forty-eight single words of various lengths (4e7

letters) were displayed (with a 6-sec time limit) in the center of

a PC screen after a fixation cross. Participants read the word

aloud as quickly as possible. Median RTs to read the words

correctly and the percentage of errors were measured.

The lexical decision task was used to evaluate the lexicality

effect. Words and nonwords were presented individually at

the center of a PC screen (for details, see Di Filippo et al., 2006).

The task required one to decide whether the stimulus was

a legal Italianword and press 1 of 2 keys as quickly as possible.

Median RTs of correct responses and percentage of errors

were measured.

The articulation task (Di Filippo et al., 2005) was used to

evaluate the contribution of articulation rate to speed of

reading aloud. The mean time (sec/digit) to perform the task

was calculated.

3.2.3. Data analysisA group of 6 healthy adults (males/females¼ 3/3), matched for

age (mean age 38.4 years, Standard Deviation e SD ¼ 1.9) and

years of education (13), constituted the control group. The

control subjects took the reading and vocal tests, aswell as the

neuropsychological tests for the general assessment without

normative Italian data: the semantic verbal fluency task,

BeeryeBuktenica developmental test of VMI, bells test, lines

cancellation task (LC), multiple features targets cancellation

task (MFCT), and letter cancellation test. Bayesian inferential

methods (Crawford and Garthwaite, 2007) were used to

compare the results of each patient with those of the controls.

Published normative data were used for the remaining tests,

and relative cutoff values used for comparison.

4. Results

4.1. Eye movements

Eyemovement traces showed oculomotor disturbancesdboth

patients had highly frequent SI, the amplitude of which was

particularly large in patient 1. Nystagmus was observed only

in patient 2. Excerpts of the traces that were recorded during

the tasks are presented in Fig. 2. As shown in the figure, both

Table 2 e Neuropsychological assessment.

Functions Tests Patient 1 Patient 2 Cutoff Controls(mean � SD)

Intellectual level WAIS-R Total IQ 91 94 <70

Verbal IQ 98 98 <70

Performance IQ 85 90 <70

Raven’s Progressive Matrices 29.4 30.3 <18.96

Visuospatial Copying drawings 8.1 7.2 <7.18

Copying drawings with landmarks 67.6 n.e. <61.85

Block design subtest (WAIS-R) 7 9 <7

Object assembly subtest (WAIS-R) 5a 6a <7

Language Token Test 34 33 <29

BADA (N of errors)

Reading Words 1/45 5/45a >2

Sentence 0/6 0/6 >2

Auditory comprehension Nouns 0/40 0/40 >2

Verbs 0/20 0/20 >2

Written comprehension Nouns 0/40 0/40 >2

Verbs 0/20 0/20 >2

Oral naming Nouns 2/30 0/30 >2

Verbs 3/28a 2/28 >2

Grammatical comprehension Auditory 1/60 0/60 >2

Visual 2/45 0/45 >2

Executive

functions

Phonemic fluency test 22.8 18.5 <17.35

Temporal rules induction 11 10 >15

Tower of London planning task 32 34 <29.1

Verbal memory Rey’s 15 mots Short term 43.8 47.8 <28.53

Digit span subtest (WAIS-R) Long term 8.6 11.1 <4.69

Forward 6 6 <5

Backward 5 5 <3

Visuospatial

memory

Immediate Visual memory 18.5 20.6 <13.85

Corsi test Forward 6 5 <5

Backward 6 6 <3

Visuomotor

integration

VMI VMI N of items 16* 11x 23.8 � 2.2

Visual perception test N of items 25 18** 24.7 � 1.9

Motor coordination test N of items 10xx 8xxx 25.7 � 1.4

Attention TAP Tonic alertness RT (msec) 473a 294 5th percentile: 327

Fasic alertness RT (msec) 426a 253 5th percentile: 307

Alertness (Tonic þ Fasic) Omissions 0/80 0/80

Anticipations 0/80 0/80

Divided attention RT (msec) 821a 812a 5th percentile: 735

Omissions 2/32 1/32

False reactions 2/32 1/32

Anticipations 0/80 0/80

Go/no-go RT (msec) 766a 607 5th percentile: 638

Omissions 1/24 0/32

False reactions 0/24 0/32

Anticipations 0/24 0/80

Line cancellation Omissions 0 0 0 � 0

Time (s) 70*** 90x 30.5 � 9.01

MFTC Omissions 4* 3* .50 � .84

Time (s) 45* 40* 56.5 � 21.5

Letter cancellation Omissions 6** 0 .60 � 1.34

Time (s) 155 240** 104.8 � 28.6

Bells test Omissions 17xxx 10x 1.00 � 1.26

Rapidity 40 29 71.8 � 27.4

WAIS-R (Orsini and Laicardi, 2001; Wechsler, 1997); Raven’s Progressive Matrices (Raven, 1947); Copying drawings with or without landmarks

(Gainotti et al., 1977); BADA (Miceli et al., 1994); Token test (De Renzi and Faglioni, 1978; De Renzi and Vignolo, 1962); Verbal fluency tasks

(Borkowsky et al., 1967); Temporal rules induction task (Villa et al., 1990); Tower of London planning task (Shallice, 1982); Rey’s 15 Words Test

(Rey, 1958); Immediate visualmemory task (Carlesimo et al., 1996); Corsi Test (Corsi, 1972); VMI¼ BeeryeBuktenica developmental test of Visual-

Motor Integration (Beery, 1997; Preda, 2000); TAP (Zimmermann and Fimm, 1995); Bells test (Gauthier et al., 1989); Line cancellation and MFTC

(Gainotti et al., 2001); Letter cancellation test (Weintraub and Mesulam, 1985). n.e. ¼ not evaluable. *p < .05; **p < .01; ***p < .005; xp < .001;

xxp < .0001; xxxp < .00005.

a Performance below the lower limit of reference data.

c o r t e x 4 9 ( 2 0 1 3 ) 6 9 1e7 0 1 695

Fig. 2 e Excerpts of eye movement traces in the healthy control, patient 1, and patient 2. Black line represents gaze position

as a function of time. Gray line in insets a and b represents target position. (a) Fixation task: note the vertical displacements

in the gaze traces of patient 1 and patient 2, indicative of SI. Control data were comparable with reference data (Abadi and

c o r t e x 4 9 ( 2 0 1 3 ) 6 9 1e7 0 1696

c o r t e x 4 9 ( 2 0 1 3 ) 6 9 1e7 0 1 697

patients presented clear pathological features. To provide

a direct comparison, patients’ results were compared to those

of the healthy sibling that was representative of the healthy

population’s performances.

4.1.1. Fixation task (Fig. 2a)The severe condition of both patients is clearly shown: their

discontinuous fixation profile is contrasted with the normal

profile of steady fixation over time performed by the healthy

sibling. SI frequency, amplitude, and duration of the sibling

(25.5� .7 SI per minute, 1.01� .68�, and 283� 106 msec,

respectively) were comparable with the reference data from

a healthy population (18.0� 14.3 SI per minute, .6� .5�,225� 150 msec, Abadi and Gowen, 2004). Both patients had

difficulties in maintaining steady fixationddue to frequent

and very large SI in patient 1 and both SI and nystagmus in

patient 2. Patient 1 performance was more than 4 SDs below

the reference data for both frequency and amplitude of

intrusions, and showed continuous square wave oscillations,

alternating biphasic square wave intrusions (BSWIs), and

monophasic square wave intrusions (MSWIs). Patient 2

exhibited MSWIs that were below the SI reference data of

more than 2 SDs for frequency and of more than 4 SDs for

amplitude.

Mean frequency, amplitude, velocity, and duration of SI are

shown in Table 3.

SI were significantly more frequent in patients than in the

control. Patient 1 had a significantly higher rate of SI than

patient 2. SI were significantly larger in patient 1 and patient 2

versus the control, and larger in patient 1 versus patient 2

(p< .00001). Moreover, mean velocity was significantly higher

in patient 1 and patient 2 compared with the control. SI

duration in the patients did not differ from the control data.

4.1.2. Saccade task (Fig. 2b)Despite the differences in trace appearance, the physiological

staircase pattern was preserved in both patients. Differences

from the control were due primarily to saccadic oscillations in

patient 1 and nystagmus in patient 2. Response latencies are

shown in Table 3. Total latency and latencies during rightward

and leftward sequences were significantly longer in the

patients than in the control but comparable between the 2

subjects. Leftward sequence latencies were longer than

rightward latencies in patient 1. No directional bias existed in

patient 2 or the control.

4.1.3. Reading task (Fig. 2c)The control’s reading pattern and fixation data (1.5 fixations

per word and 238 msec mean fixation duration for short

words) were consistent with values reported for healthy

population (mean number of fixations without constraint on

word length: ca. 1.3; average fixation duration: 200e250 msec;

Gowen, 2004). (b) Saccade task: note that the staircase pattern (

oscillations in patient 1 and by nystagmus in patient 2. (c) Shor

staircase shape of reading (upward displacement representing r

horizontal displacement indicative of fixation, and large downw

return to the next line of text) in the control gaze trace was distu

2. Similar patterns were detected for long words and short and

e.g., Starr and Rayner, 2001). The characteristic rightward

staircase pattern of normal reading was present in all partic-

ipants, although it was masked by SI or nystagmus in the

patients’ traces.

The number of fixations and their durations are shown in

Table 3. The mean number of fixations per item and mean

fixation duration in both patients did not significantly differ

from that of the healthy control under any stimulus.

4.2. Neuropsychological profile

4.2.1. Patient 14.2.1.1. NEUROPSYCHOLOGICAL EXAMINATION. Patient 1’s neuro-

psychological test results are listed in Table 2. Her intellectual

level was within the normal range. Although she showed no

sign of constructional apraxia, she received below-normal

scores on the object assembly WAIS-R (Wechsler Adult Intel-

ligence Scale Revised) subtest. Executive functions, verbal and

spatial memory, and language abilities were preserved.

Within the VMI, she performed worse than the control

subjects on the VMI and motor coordination subtests.

In the LC test, she showed normal scanning strategy, albeit

with significantly longer execution times than controls. In

more complex tasks, such as the MFTC, Letters cancellation

test, and Bells test, patient 1’s performance differed signifi-

cantly from those of controls.

Patient 1 was accurate on all subtests of the test for

attentional performance (TAP), albeit extremely slow. Also,

the index of phasic alertness was good (Zimmermann and

Fimm, 1995).

In summary, patient 1 showed motor and VMI difficulties,

with good although slow performance onmost cognitive tasks.

4.2.1.2. ADDITIONAL READING AND VOCAL TESTS. On the nDR task,

patient 1’s reading times were longer than the controls’ under

all conditions (Table 4).

Her reading time was greater for nonwords than words,

a difference that was comparable with that of controls for

both vertical and horizontal displays. Words were read more

accurately than nonwords; this difference was significantly

greater in patient 1 than in the control group for vertical and

horizontal displays.

On the simple vocal reaction time task (Fig. 3a), patient 1 had

slower vocal RTs than the controls under uncued

(median ¼ 681 msec) and cued (median ¼ 659 msec) condi-

tions. The difference between the conditions (22 msec) did not

differ significantly versus the control group (45 msec), indi-

cating that her ability to take advantage of the cues was

preserved.

On the vocal reaction time to 4e7-letter words (Fig. 3a), patient

1 respondedmore slowly than controls. RTs did not depend on

evident in the control’s trace) was disrupted by saccadic

t words reading task: note that the typical rightward

ightward movements for reading progression, followed by

ard displacement indicative of leftward movements to

rbed by intrusions in patient 1 and by nystagmus in patient

long nonwords (not shown here).

Table 3e Eyemovement parameters in the Fixation, Saccade, and Reading tasks.Mean values (with SD in parentheses) arereported for both patients and the healthy control. Probability values for t-test comparisons between each patient and thecontrol are also reported.

Patient 1 Patient 2 Control

Mean (SD) p-level Mean (SD) p-level Mean (SD)

Fixation task

Frequency of SI (per minute) 81.0 (17.0) <.00001 52.0 (19.3) <.00001 25.5 (.7)

SI amplitude (deg) 5.37 (4.09) <.00001 2.96 (3.37) <.005 1.01 (.68)

SI velocity (deg/sec) 154.3 (66.2) <.05 86.4 (62.0) <.05 55.3 (23.5)

Fixation duration following SI (msec) 243 (115) n.s. 300 (134) n.s. 283 (106)

Saccade task

Total saccadic latency (msec) 249 (67) <.00001 238 (41) <.00001 145 (27)

Rightward sequences latency (msec) 205 (49) <.0001 229 (35) <.0001 142 (29)

Leftward sequences latency (msec) 288 (57) <.0001 250 (48) <.0001 149 (26)

Left -right seq. latency difference (msec) 83 21 7

Reading task

Mean number of fixations per item

Short words 1.5 (.2) n.s. 1.8 (.4) n.s. 1.5 (.4)

Long words 1.6 (.4) n.s. 1.8 (.5) n.s. 1.8 (.4)

Short nonwords 1.6 (.4) n.s. 2.1 (.3) n.s. 1.9 (.4)

Long nonwords 2.2 (.4) n.s. 2.3 (.4) n.s. 2.4 (.5)

Mean fixation duration (msec)

Short words 257 (164) n.s. 291 (124) n.s. 238 (96)

Long words 258 (222) n.s. 284 (110) n.s. 239 (82)

Short nonwords 261 (137) n.s. 283 (214) n.s. 260 (78)

Long nonwords 268 (163) n.s. 300 (186) n.s. 263 (73)

c o r t e x 4 9 ( 2 0 1 3 ) 6 9 1e7 0 1698

word length (t-test between pairs of conditions: t< 1 each,

n.s.). The percentage of naming errors was 0%.

Onthe lexical decision task (Fig. 3b), patient1hadslowervocal

RTs than controls under all conditions. The latency difference

between nonwords and words (nonwords minus words: 189

and 332 msec for 3- and 5-letter items, respectively) was not

significantly greater than in controls (46 and 186 msec), indi-

cating no specific impairment with nonwords. Her accuracy

was comparable with that of controls under all conditions.

On the articulation task, the mean time to count numbers

(.86 sec/digit) was comparable with that of controls (mean

.86 sec/digit, SD ¼ .09).

Table 4 e nDR task results.

Tasks

Pat

Vertical display Words 1

Nonwords 1

WeNW difference

Horizontal display Words

Nonwords 1

WeNW difference

Vertical display Words 1

Nonwords 11

WeNW difference 10

Horizontal display Words

Nonwords 7

WeNW difference 7

*p < .05; **p < .01; ***p < .005; xp < .001.

In summary, when reading lists, patient 1 made a higher

percentage of errors than controls only for nonwords. Laten-

cies were longer on all RTs tasks, independent of stimulus.

However, when displayed one by one, the ability to read

nonwords was not specifically impaired with regard to laten-

cies and accuracy. Phasic alertness and articulation rate were

maintained.

4.2.2. Patient 24.2.2.1. NEUROPSYCHOLOGICAL EXAMINATION. The neuro-

psychological test results of patient 2 are listed in Table 2.

Patient 2’s intellectual level was within the normal range. No

Time in sec per item

ient 1 Patient 2 Controls (Mean ± SD)

.09x 1.04*** .57 � .09

.74* 1.97* 1.07 � .26

.65 .93* .5 � .19

.98x .98x .54 � .07

.29* 1.16 .77 � .21

.31 .18 .24 � .14

Percentage of errors

.09 .00 1.27 � 1.44

.96* 8.70 3.98 � 2.81

.87* 8.7 2.72 � 3.49

.00 1.82 1.36 � 1.60

.27* 9.09** 3.49 � 1.56

.27* 7.27* 2.12 � 1.88

Fig. 3 e a) Median vocal RTs as a function of alert condition on the Simple vocal reaction time task and as a function of word

length on the Vocal RTs to 4e7-letter words task. (b) Median RTs as a function of lexicality (words or nonwords) and length

(3- or 5-letter items) on the Lexical decision task. For comparison, mean RTs for the control group are reported. Error bars

indicate SDs. Asterisks indicate level of statistical significance as follows: (*) .05 < p < .07; *p < .05; **p < .01; ***p < .005.

c o r t e x 4 9 ( 2 0 1 3 ) 6 9 1e7 0 1 699

sign of constructional apraxia was present; however, he per-

formed poorly on the object assembly WAIS-R subtest.

Although patient 2 made 5 errors on the battery for the

analysis of aphasic disorders (BADA) word reading subtest,

language ability, executive functions and verbal and spatial

memory were intact. Conversely, the functions that were

tested on all VMI subtests were impaired.

Patient 2 performed well on the lines and letters cancella-

tion tests but had slow scanning times. He made errors on the

MFTC and bells test. Overall, scanning strategies appeared to

be structured little.

Patient 2 was accurate but slow on all TAP subtests,

although he scored below the 5th percentile only on the

divided attention subtest. The index of phasic alertness was

normal.

Overall, patient 2 presented with reduced VMI and visuo-

spatial attention. RTs were within the normal range on tasks

with low attentional or motor demand but were delayed on

the most demanding tasks.

4.2.2.2. READING AND VOCAL TESTS. On the nDR task, reading times

were longer than for thecontrolsunder all conditionsexcept for

horizontally presented nonwords (Table 4). Reading time was

greater for nonwords than words, a difference that was

comparable with that of controls for the horizontal but not

vertical display. Errors were more frequent than in controls on

the horizontal nonwords task. Words were read more accu-

rately than nonwords, but the difference was significantly

greater than in thecontrol grouponly for thehorizontaldisplay.

On the simple vocal reaction time (Fig. 3a), RTs did not

differ significantly from those of controls under uncued

(median ¼ 504 msec) or cued (median ¼ 452 msec) conditions.

The difference between conditions (52 msec) was comparable

with that of the control group (45 msec).

On the vocal reaction time to 4e7-letter words (Fig. 3a), patient

2 had slower RTs than controls for all word lengths. Latency

differences as a function of word length were not significant

(t-test between pairs of conditions: t< 1 each, n.s.). The

percentage of naming errors was 0%.

On the Lexical decision task (Fig. 3b), RTs were significantly

slower than controls for 5-letter words; slowness approached

significance for 3-letter words and nonwords. The latency

difference between nonwords and words (nonwords minus

words: 125 and 153 msec for 3- and 5-letter items, respec-

tively) was not significantly greater than in controls (46 and

186 msec). Accuracy was comparable with that of controls

under all conditions.

On the articulation task, the mean time (.87 sec/digit) was

comparablewith that of controls (mean .86 sec/digit, SD¼ .09).

In summary, compared with controls, patient 2

committed a higher percentage of errors for horizontally

presented nonwords and had slower reading times for

vertically presented nonwords. Latencies were slower for

complex stimuli (e.g., words) but not simple stimuli (simple

vocal RT). The ability to read nonwords was not specifically

impaired with regard to latency or accuracy when shown one

by one.

5. Discussion

In this study, eye movement recordings and cognitive tasks

were used to examine oculomotor and cognitive deficits in 2

siblings with AOA2. The patients presented with similar

c o r t e x 4 9 ( 2 0 1 3 ) 6 9 1e7 0 1700

general clinical conditions, corresponding to the hallmarks of

AOA2 (Anheim et al., 2009; Le Ber et al., 2004), with few

differences between them with regard to oculomotor and

cognitive symptoms.

Notably, the accurate oculomotor screen allowed us to

detect new features of the oculomotor pattern in AOA2. Based

on eye movement recordings, we observed frequent SI on all

tasks in both patients, including reading. SI are common in

the normal population (Abadi and Gowen, 2004). Abnormal SI

have reported in several neurological diseases (Fielding et al.,

2006; Rascol et al., 1991) but have never been described in

AOA2 (Le Ber et al., 2004). In this study, both siblings experi-

enced increased frequencies of SI. Moreover, the latency of

visually guided saccades using a highly predictable target was

longer.

SI and/or nystagmus disrupted the performance on word

and nonword reading tasks. Despite the repeated interrup-

tions due to SI, the staircase pattern that is typical of reading

remained in both patients (when netted from SI and

nystagmus). Overall eye movements were similarly affected

on nonverbal and verbal tasks. The presence of intrusions or

nystagmus caused the slow performance on all reading tests.

To determine the impact of oculomotor impairments on

reading, we used a specific task (nDR task) that separates the

effects between the positional and lexical/semantic charac-

teristics of stimuli. In this task, words and nonwords were

presented in lists, resulting in a more complex visual array

than single-item presentations. In fact, when items were

presented individually to the patients, the lexicality effect was

maintained and they did not make any errors despite being

slower than controls. Errors appeared when the display of

items enhanced oculomotor requests (i.e., a list). In this case

the lexicality effect was confirmed, indicating that the

semantic process was preserved despite difficulties with eye

scanning. Thus, while reading scanning was attenuated by

oculomotor impairments, cognitive processingwas preserved,

demonstrating that the reading disabilities were caused by

oculomotor deficits, not cognitive problems.

With regard to the neuropsychological profiles, very few

deficits were detected, other than the reading difficulties that

we have described, which primarily involved visuomotor

integration and attention.

The RTs on tasks that demanded attention and motor

responses (manual or vocal) to visual stimuli (both simple and

complex) were slow in patient 1 but preserved in patient 2.

Pathological latencies were always observed for patient 1,

whereas the pattern of response in patient 2 fell onlywhen the

visual-attention or visuomotor load was heavierdi.e., when

decision-making was required or multiple and complex visual

stimuli were shown. Thus, whereas patient 2 had subtle signs

of attention problems, we could not draw conclusions on the

attention of patient 1 due to her general slowness.

In general, our oculomotor and cognitive assessments

indicated that both AOA2 patients had substantially preserved

neuropsychological profiles with deficits whose relevance

must be carefully evaluated due to the concomitant oculo-

motor problems. Most problems in the cognitive domain were

consistent with oculomotor difficulties, a pattern that is

consistent with the existence of vermal atrophy in both

patients. Cerebellar oculomotor control resides in vermal

structures (Glickstein et al., 2011), and cerebellar modulation

of cognition depends on the hemisphere (Stoodley and

Schmahmann, 2009). Patient 1, who had more diffused cere-

bellar atrophy, presented consistently with general slowness

on all tasks. The deficits were less severe in patient 2, whose

atrophy was more limited and mostly evident in the vermis.

Mild cognitive deficits in AOA2 have been reported by Le

Ber et al. (2004) and by the present study reporting attention

deficits and reading difficulties in two patients. These findings

are consistent with the high variability of clinical expression

in AOA2 (Anheim et al., 2009; Le Ber et al., 2004).

Unlike Le Ber et al. (2004),wedidn’t find executive deficits. It

is worth noting that this discrepancy may be due to the

different assessment performed to analyze executive func-

tions. Indeed, we only used standard non-computerized tests.

An antisaccade task, that is not performed in the present

study, could have been useful to reveal prefrontal dysfunction.

Several studies addressed the issue of genotype/phenotype

correlations with no clear-cut conclusions (Moreira and

Koenig, 2009). In particular no significant difference, with

respect to ocular motility, has been found comparing patients

carrying mutations within and outside the helicase domain

(Anheim et al., 2009).

In summary, our findings indicate that in patients with

a complex clinical syndrome, such as AOA2, the performance

on neuropsychological tests, especially those that require

rapid performance and eye or hand-eye control, must be

analyzed with respect to the oculomotor components. The

possible influence of attentional disorders on cognitive

performance should be analyzed carefully.

Acknowledgments

We thank patient 1 and patient 2 for the hours of testing that

they patiently endured. The editing support of Blue Pencil

Science is also acknowledged. This work was supported in

part by grants to Marco Molinari and Maria G. Leggio from the

Italian Ministry of Instruction, University and Research and

the Italian Ministry of Health.

r e f e r e n c e s

Abadi RV and Gowen E. Characteristics of saccadic intrusions.Vision Research, 44(23): 2675e2690, 2004.

Anheim M, Monga B, Fleury M, Charles P, Barbot C, Salih M, et al.Ataxia with oculomotor apraxia type 2: Clinical, biological andgenotype/phenotype correlation study of a cohort of 90patients. Brain, 132(10): 2688e2698, 2009.

Beery K. The BeeryeBuktenica Developmental Test of Visual-motorIntegration: VMI with Supplemental Developmental Tests of VisualPerception and Motor Coordination: Administration, Scoring andTeaching Manual. Parsippany, NY: Modern Curriculum Press,1997.

Borkowsky JG, Benton AL, and Spreen O. Word fluency and braindamage. Neuropsychologia, 5: 135e140, 1967.

Carlesimo G, Caltagirone C, and Gainotti G. The mentaldeterioration battery: Normative data, diagnostic reliabilityand quantitative analyses of cognitive impairment. EuropeanNeurology, 36(6): 378e384, 1996.

c o r t e x 4 9 ( 2 0 1 3 ) 6 9 1e7 0 1 701

Corsi PM. Human memory and medial temporal region of thebrain. [Dissertation Abstracts International] McGill University(University Microfilms No. AAI05-77717), 34(02), 891B, 1972.

Crawford JR and Garthwaite PH. Comparison of a single case toa control or normative sample in neuropsychology:Development of a Bayesian approach. CognitiveNeuropsychology, 24(4): 343e372, 2007.

Criscuolo C, Chessa L, Di Giandomenico S, Mancini P, Sacca F,Grieco GS, et al. Ataxia with oculomotor apraxia type 2: Aclinical, pathologic, and genetic study. Neurology, 66(8):1207e1210, 2006.

De Luca M, Di Pace E, Judica A, Spinelli D, and Zoccolotti P. Eyemovement patterns in linguistic and non-linguistic tasks indevelopmental surface dyslexia. Neuropsychologia, 37(12):1407e1420, 1999.

De Luca M, Borrelli M, Judica A, Spinelli D, and Zoccolotti P.Reading words and pseudo-words. An eye movement study ofdevelopmental dyslexia. Brain & Language, 80(3):617e626, 2002.

De Renzi E and Faglioni P. Normative data and screening power ofa shortened version of the Token Test. Cortex, 14(1):41e49, 1978.

De Renzi E and Vignolo LA. The Token Test: A sensitive test todetect receptive disturbances in aphasics. Brain, 85(4):665e678, 1962.

Di Filippo G, Brizzolara D, Chilosi A, De Luca M, Judica A, Pecini C,et al. Rapid naming, not cancellation speed or articulationrate, predicts reading in an orthographically regular language(Italian). Child Neuropsychology, 11(4): 349e361, 2005.

Di Filippo G, De Luca M, Judica A, Spinelli D, and Zoccolotti P.Lexicality and stimulus length effects in Italian dyslexics: Roleof the overadditivity effect. Child Neuropsychology, 12(2):141e149, 2006.

Fielding J, Georgiou-Karistianis N, Bradshaw J, Millist L,Churchyard A, and White O. Accelerated time-course ofinhibition of return in Huntington’s disease. Behavioural BrainResearch, 166: 211e219, 2006.

Gainotti G, Miceli G, and Caltagirone C. Constructional apraxia inleft brain-damaged patients: A planning disorder? Cortex,13(2): 109e118, 1977.

Gainotti G, Marra C, and Villa G. A double dissociation betweenaccuracy and time of execution on attentional tasks inAlzheimer’s disease and multi-infarct Dementia. Brain, 124(4):731e738, 2001.

Gauthier L, Dehaut F, and Joanette Y. The bells test: Aquantitative and qualitative test for visual neglect.International Journal of Clinical Neuropsychology, 11: 49e53, 1989.

Glickstein M, Sultan F, and Voogd J. Functional localization in thecerebellum. Cortex, 47(1): 59e80, 2011.

Le Ber I, Bouslam N, Rivaud-Pechoux S, Guimaraes J, Benomar A,Chamayou C, et al. Frequency and phenotypic spectrum ofataxia with oculomotor apraxia 2: A clinical and genetic studyin 18 patients. Brain, 21(4): 1e9, 2004.

Leggio MG, Chiricozzi FR, Clausi S, Tedesco AM, and Molinari M.The neuropsychological profile of cerebellar damage: Thesequencing hypothesis. Cortex, 47(1): 137e144, 2011.

Miceli G, Laudanna A, Burani C, and Capasso R. Batteria perl’analisi dei deficit afasici. Cepsag, Roma: BADA, 1994.

Moreira MC, Klur S, Watanabe M, Nemeth AH, Le Ber I, Moniz JC,et al. Senataxin, the ortholog of a yeast RNA helicase, ismutantin ataxia-ocular apraxia 2. Nature Genetics, 36: 225e227, 2004.

Moreira MC, Barbot C, Tachi N, Kozuka N, Uchida E, Gibson T,et al. The gene mutated in ataxia-ocular apraxia 1 encodes thenew HIT/Zn-finger protein aprataxin. Nature Genetics, 29:189e193, 2001.

Moreira MC and Koenig M. Ataxia with oculomotor apraxia type 2.In Pagon RA, Bird TD, Dolan CR, Stephens K, (Eds),GeneReviews [Internet]. Seattle, WA: University ofWashington, 1993e2004 [updated 2009].

Orsini A and Laicardi C. Wais-R. Contributo alla taratura italiana.Firenze: Organizzazioni Speciali, 2001.

Porac C and Coren S. Lateral Preferences and Human Behaviour.New York: Springer-Verlag, USA, 1981.

Preda C. “Beery-Buktenica” con i test supplementari Percezione Visiva eCoordinazione motoria. Firenze: Organizzazioni Speciali, 2000.

Raven JC. Progressive Matrices. Sets A, Ab, B: Board and Book Forms.London: Lewis, 1947.

Rascol O, Sabatini U, Simonetta-Moreau M, Montastruc JL,Rascol A, and Clanet M. Square wave jerks in Parkinsoniansyndromes. Journal of Neurology, Neurosurgery & Psychiatry,54(7): 599e602, 1991.

Rey A. Memorisation de serie de 15 mots en 5 repetitions. In Rey A(Ed), L’Examen Clinique en Psychologie. Paris: Universiteries deFrance Press, 1958.

Sartori G, Job R, and Tressoldi E. Batteria per la Valutazione dellaDislessia e della Disortografia Evolutiva. Firenze: OrganizzazioniSpeciali, 1995.

Shallice T. Specific impairments in planning. PhilosophicalTransactions of the Royal Society B Biological Sciences, 289(1089):199e209, 1982.

Spinelli D, De Luca M, Judica A, and Zoccolotti P. Crowding effectson word identification in developmental dyslexia. Cortex, 38(2):179e200, 2002.

Starr MS and Rayner K. Eye movements during reading: Somecurrent controversies. Trends in Cognitive Science, 5(4):156e163, 2001.

Stoodley CJ and Schmahmann JD. Functional topography in thehuman cerebellum: A meta-analysis of neuroimaging studies.NeuroImage, 44(2): 489e501, 2009.

Tazir M, Ali-Pacha L, M’Zahem A, Delaunoy JP, Fritsch M,Nouioua S, et al. Ataxia with oculomotor apraxia type 2: Aclinical and genetic study of 19 patients. Journal of theNeurological Sciences, 278(1e2): 77e81, 2009.

Trouillas P, Takayanagi T, Hallett M, Currier RD, Subramony SH,Wessel K, et al. International Cooperative Ataxia Rating Scalefor pharmacological assessment of the cerebellar syndrome.Journal of the Neurological Sciences, 145(2): 205e211, 1997.

Villa G, Gainotti G, De Bonis C, and Marra C. Double dissociationbetween temporal and spatial pattern processing in patientswith frontal and parietal damage. Cortex, 26(3): 399e407, 1990.

Wechsler D. Wais-R Wechsler Adult Intelligence Scale Revised.Firenze: Organizzazioni Speciali, 1997.

Weintraub S and Mesulam MM. Mental state assessment ofyoung and elderly adults in behavioural neurology. InMesulam MM (Ed), Principles of Behavioural Neurology.Philadelphia: Devis Company, 1985: 71e123.

Zimmermann P and Fimm B. Testbatterie zurAufmerksamkeitsprufung (TAP) Test Battery for the Assessment ofAttentional Dysfunction. Herzogenrath, Germany: Psytest, 1995.

Zoccolotti P, De Luca M, Judica A, and Burani C. Delayed namingcancels the word length effect in developmental dyslexia.Brain & Language, 99: 36e37, 2006.