Brain and Cognition 72 (2010) 337–346

Contents lists available at ScienceDirect

Brain and Cognition

journal homepage: www.elsevier .com/ locate /b&c

Impaired awareness of movement disorders in Parkinson’s disease

Martina Amanzio a,f,*, Silvia Monteverdi a, Alessandra Giordano b, Paola Soliveri c, Paola Filippi d,e,Giuliano Geminiani a,f

a Department of Psychology, University of Torino, Via Verdi 10, 10123 Torino, Italyb Department of Neuroscience, University of Torino, Italyc Istituto Nazionale Neurologico ‘‘Carlo Besta”, Department of Neurology I Milano, Italyd Azienda Sanitaria Ospedaliera San Luigi Gonzaga, Neurology Unit, Orbassano, Torino, Italye Ospedale Maria Vittoria, Neurology Unit, Torino, Italyf Neuroscience Institute of Turin, NIT, University of Torino, Italy

a r t i c l e i n f o

Article history:Accepted 19 October 2009Available online 14 November 2009

Keywords:Parkinson’s diseaseAwareness of movement disordersDyskinesiaHypokinesiaExecutive functions

0278-2626/$ - see front matter � 2009 Elsevier Inc. Adoi:10.1016/j.bandc.2009.10.011

* Corresponding author. Address: Department oTorino, Via Verdi 10, 10123 Torino, Italy. Fax: +39 11

E-mail address: amanzio@psych.unito.it (M. Aman

a b s t r a c t

Background: This study analyzed the presence of awareness of movement disorders (dyskinesias andhypokinesias) in 25 patients with Parkinson’s disease (PD) and motor fluctuations (dyskinesias, wearingoff, on–off fluctuations). Of the few studies that have dealt with this topic, none have analyzed the differ-ences in the awareness of motor deficits by comparing the on and off states using motor scales and anextensive battery of tests to assess cognitive and behavioral functioning.Methods: PD patients were compared on three different scales that we have devised to measure aware-ness of movement disorders: Global Awareness of Movement (GAM) Disorders, dyskinesia/hypo-brady-kinesia rating scales.Results: Data showed that PD patients had greater awareness and psychological suffering in the off statethan in the on state. In particular, they were troubled by motor disabilities related to hypokinesias andhad mood-related symptoms and a perception of disability in activities of daily living. Interestingly,patients only showed a selective reduction of awareness of movement disorders associated with execu-tive functions and related to dyskinesias in the on state, compared to a preserved awareness of hypoki-nesias in the off state. On the contrary, no association with executive functions was found in the off state.Conclusion: Our findings suggest that the dopaminergic overstimulation of mesocorticolimbic pathwaysmay cause a dysfunction of prefrontal–subcortical connections related to the impaired insight.

� 2009 Elsevier Inc. All rights reserved.

1. Introduction dopamine) is complicated by motor fluctuations with dyskinesias

Parkinson’s disease (PD) is a progressive, neurodegenerativedisorder, characterized by complex motor impairment with bothhypokinetic (hypo-bradykinesia) and hyperkinetic (resting tremor)symptoms. The neuropathological hallmark of PD is the degenera-tion of dopaminergic neurons of the substantia nigra pars com-pacta (Jellinger, 1987), leading to dopaminergic denervation ofthe striatum, which is the first aspect to occur. Subsequently, asthe disease progresses, other dopaminergic systems are involved;in particular, the mesocorticolimbic dopaminergic depletion dif-fuses to the ventral striatum (Agid et al., 1993; Kish, Shannak, &Hornykiewicz, 1988) producing mild executive function deficitsand behavioral alterations (Owen et al., 1992; Taylor, Saint-Cyr,& Lang, 1986). The role of dopaminergic treatment on motor,behavioral and cognitive dysfunctions is complex. From a motorperspective, the efficacy of treatment with levodopa (precursor of

ll rights reserved.

f Psychology, University of6702061.zio).

and wearing off or on–off states.Although PD is characterized by a complex and changeable

spectrum of symptoms, the subjective perception of motor impair-ment is an interesting phenomenon that has been inadequatelyanalyzed. Reduced awareness of neurological symptoms, definedas ‘‘anosognosia”, can involve a wide domain of situations (Wein-stein & Kahn, 1950). In particular, in the domain of motor functionsthe most widely investigated symptoms are anosognosia for hemi-plegia and anosognosia for dyskinetic movements (Myslobodsky,1986; Shenker, Wylie, Fuchs, Manning, & Heilman, 2004). The phe-nomenon of anosognosia could be considered in terms of at leasttwo different theoretical frames. For the motor-sensorial domainsof anosognosia, such as hemiplegia, an impairment of a modularsystem of awareness can be hypothesized (Bisiach & Geminiani,1991). On the other side, prefrontal–striatal dysfunction of theexecutive monitoring system could have a role in other domainsof reduced awareness (McGlynn & Schacter, 1989). For example,it has been suggested that a dysfunction of the Central ExecutiveSystem (Baddeley, 1986) may account for the lack of awarenessin Alzheimer’s disease (AD) patients (Amanzio & Torta, 2009;

338 M. Amanzio et al. / Brain and Cognition 72 (2010) 337–346

Lopez, Becker, Somsak, Dew, & DeKosky, 1994). This second typeof unawareness due to executive dysfunctions has not yet beenreported in patients with PD.

Unawareness of deficits appears to be associated not only withdamage to cortical brain regions (McGlynn & Kaszniak, 1991;Starkstein et al., 1996), but also with subcortical damage (Godef-roy, Rousseaux, Pruvo, Cabaret, & Leys, 1994; Healton, Navarro,Bressman, & Brust, 1982; Jacome, 1986), in particular anosognosiaof dyskinetic movements (Lazzarino & Nicolai, 1991). Few studieshave examined the role of unawareness of deficits in PD patients.Starkstein et al. (1996), comparing AD and PD patients on an exten-sive battery of neuropsychological and psychiatric measures, founda higher level of deficit unawareness among AD patients than PDpatients and a disinhibitive syndrome, showing in the AD popula-tion a frontotemporal dysfunction of cortical structures. Anotherstudy evaluated the unawareness of dyskinesias in PD and Hun-tington’s disease (HD) patients and found that both groups wereimpaired in detecting the presence of their dyskinetic movements,suggesting that this aspect of anosognosia could be related to sub-cortical dysfunction (Vitale et al., 2001). In particular, in PD pa-tients the level of unawareness was inversely related to theseverity of the dyskinesias, while in HD patients it was directly re-lated to the duration of the disease. However, this study had someimportant limitations: the parkinsonian patients were not studiedon the basis of a neuropsychological and neuropsychiatric assess-ment, cognitive deterioration was not excluded and no measureof awareness of hypokinetic disorders was considered. Anotherstudy (Seltzer, Vasterling, Mathias, & Brennan, 2001), comparingAD and PD patients, showed that the unawareness exhibited bythe PD group was more strongly related to neuropsychological dys-functions, in terms of poor overall cognitive function, especiallythose assessing memory; in this direction, the authors underlinedthat PD patients with intact cognitive functions display relativelypreserved awareness of motor deficits. In particular, this studydemonstrated a correlation between the attention subscale of theDementia Rating Scale and the discrepancy between patients’and caregivers’ evaluations of motor deficit; however, as theauthors themselves admit, this can only be considered as weakconfirmation of the link between this type of awareness and exec-utive dysfunctions. It is important to underline that the study bySeltzer et al. (2001) did not analyze differences in awareness ofmotor deficits comparing on and off states and neuropsychologicaland neuropsychiatric variables in the same patients. Finally, a morerecent study (Leritz, Loftis, Crucian, Friedman, & Bowers, 2004)investigating self-awareness of non-demented PD patients showeddiscrepancies between patients’ and caregivers’ reports on auton-omy in activities of daily living (IADL). In particular, patients de-scribed themselves as less impaired compared to caregivers’evaluations of the patients’ level of disability. This effect was morepronounced in patients with left-side motor symptoms (greaterright basal ganglia dysfunction); the authors concluded that basalganglia dysfunction might alter insight into the severity of illnessmore prevalently in patients with right hemisphere lesions. How-ever, even this study did not analyze unawareness of motor deficitsby differentiating between motor fluctuations in the on and offstates.

From the above considerations, an analysis of the connectionbetween the frontal lobes and the basal ganglia would appear tobe a useful model for explaining the presence of unawareness ofdeficits in subcortical cognitively intact PD patients. With thisaim we analyzed the awareness of movement disorders relatedto pharmacological therapy in the on and off states. We thus ex-pected to find a reduced awareness of hyperkinetic deficits in theon state, associated with the detrimental role of dopaminergictreatment on the prefrontal–subcortical loops producing executivedisabilities and a preserved awareness of hypo-bradykinesia in the

off state. We also hypothesized an association between patients’judgments concerning hypokinetic movement disorders and anxi-ety-depression mood orientation, attesting a preserved awarenessof their disabilities during the off state. Finally, we considered acomparison between PD patients’ predominantly left versus pre-dominantly right-side motor symptoms at onset of disease, to as-sess the role of this phenomenon in the awareness of motordeficits.

2. Methods

2.1. Subjects

Twenty-five patients (13 women, 12 men) with idiopathic Par-kinson’s disease, motor fluctuations (Hughes, Daniel, Kilford, &Lees, 1992) and receiving levodopa treatment (combined with car-bidopa or benserazide), often associated with dopamine agonists,were recruited to participate in the study. The demographic andclinical data of the PD population are summarized in Tables 1and 2. Patients were enrolled from a series of consecutive out-pa-tients seen at the Neurology Units of the Carlo Besta hospital(Milano, Italy) and the San Luigi Gonzaga hospital (Orbassano,Italy). The inclusion criteria were good clinical response to levo-dopa with presence of wearing off or on–off phenomena (patientswith random on–off were excluded) and peak-of-dose dyskinesias(patients with early morning and painful dystonia were alsoexcluded).

Patients were excluded from the study if they (1) had majordepression or dysthymia, based on DSM-IV criteria (1994), (2)had a Mini Mental State Examination score <24 (MMSE, Folstein,Folstein, & McHugh, 1975), (3) had a history of neurological andpsychiatric disorders (other than PD), in particular if they hadhedonistic homeostatic dysregulation, HHD (Giovannoni, O’Sulli-van, Turner, Manson, & Lees, 2000; Pezzella et al., 2003), (5) weretaking medications that could directly impact cognitive function-ing, other than dopaminergic therapy, such as antidepressants,neuroleptics and anxiolytics, (6) were unable to perform neuropsy-chological assessment in the off state.

The patients who fulfilled the above criteria were selected forinclusion in the study.

Twenty-five referring spouses and/or caregivers of the PD pop-ulation took part the study in order to provide information aboutpatients’ ability in daily living activities. Collaterals had normalneurological and psychiatric evaluations; mental deteriorationwas excluded by clinical examination and MMSE. Overall, the care-giver comparison group was demographically similar to the patientgroup based primarily on socio-economic status. Both patients andcaregivers were required to complete the North-Western Univer-sity Disability Scale, NUDS (Canter, De Latorre, & Mier, 1961), pro-viding an evaluation both in the on and off states.

Patients and caregivers participated willingly in the study andall gave their informed consent. The study was approved by theEthics Committee of the Department of Psychology, University ofTorino.

2.2. Motor, cognitive and neuropsychiatric assessment

PD patients were assessed using an extensive motor, cognitiveand psychiatric evaluation. The PD population was also analyzedin terms of side of onset with predominantly right-side or left-sidemotor symptoms.

Motor screening was performed using the Unified ParkinsonDisease Rating Scale (UPDRS, Fahn & Elton, 1987), which wasadministered by trained clinicians (neurologists) blind to the aimof the study. In particular, parkinsonian motor impairment was

Table 1Demographic, clinical, neuropsychological and neuropsychiatric measures expressed by mean (±SD). Maximum scores of the tests are shown in square parentheses.

PD patients, N = 25 Mean (±SD), Min.–Max. On state Off state

Age (years) 59.12 (8.98), 39–74Sex (f/m) 13/12Education (years) 10 (4.90), 5–18Duration of disease (months) 137.60 (42.03), 84–240Duration of motor fluctuations (months) 52.48 (36.19), 6–156Duration of levodopa pharmacological treatment (months) 114.32 (44.05), 48–228Dose of levodopa (mg per day)a 845.0 (359.92), 400–1500Unified Parkinson Disease Rating Scale (UPDRS) III* 17.36 (7.46) 46.72 (8.80)MMSE: [30] 27.80 (2.04) 27.50 (2.32)

Claridge modified test*

Total score: [9] 4.60 (3.04) 3.08 (3.03)

Wechsler Memory ScaleSubtest IV: [23] 6.76 (4.81) 6.65 (4.65)Subtest VII: [21] 11.70 (4.07) 10.65 (4.51)

WCST, modified versionTotal score: [48] 36.48 (11.61) 36.20 (12.14)Phonemic Fluency Test 30.72 (11.24) 29.92 (10.87)Brief Psychiatric Rating Scale*: [96] 29.60 (4.17) 35.76 (5.75)Hamilton anxiety scale*: [56] 17.96 (6.34) 24.96 (9.02)Hamilton depression scale*: [68] 14.04 (2.71) 19.28 (3.45)

Min.–Max.: minimum–maximum.For MMSE lower scores indicate more severe cognitive impairment.For Claridge, WMS IV and VII, WCST, Phonemic Fluency Tests higher scores indicate better performance. For the Phonemic Fluency Test the cut off is >16.For BPRS, HAM-A and HAM-D higher scores indicate more severe symptoms.Cut off values for: MMSE: 624, Claridge: 64, WCST (Nelson, 1976): >50% of perseverative errors, BPRS: >39, HAM-D: >15, HAM-A: >18.a Anti-parkinsonian medications were expressed as levodopa equivalent daily dosage (LEDD).* p < 0.03.

Table 2Detailed clinical characteristics of patients. Maximum score for UPDRS III is 56; maximum score for UPDRS-IV, part A is 13. Higher scores indicate severe symptoms.

Patientcode

Gender Age Education, durationof LD treatmenta

Duration of diseasea, durationof motor fluctuationsa

UPDRS III,on/off

UPDRSIV-part A

Hoehn andYahr, on/off

Side ofonset

Medication and dosage

1 M 66 17, 60 120, 12 16/41 4 4/3 Right 600 mg levodopa; 1.5 mg pramipexole2 F 55 13, 84 96, 6 12/40 4 5/4 Left 1000 mg levodopa; 20 mg ropinirole3 M 68 13, 72 96, 60 15/48 4 5/4 Left 1300 mg levodopa; 3 mg pramipexole4 F 64 5, 228 240, 156 20/56 6 5/4 Left 400 mg levodopa; 3 mg pramipexole5 M 59 15, 72 120, 36 14/52 6 4/2.5 Right 1000 mg levodopa; 20 mg ropinirole6 M 63 13, 119 120, 108 15/43 3 3/2 Right 1500 mg levodopa; 4.5 mg pramipexole7 M 68 17, 108 228, 12 24/43 5 4/3 Right 1149 mg levodopa8 M 74 8, 204 204, 36 19/53 6 5/4 Right 1400 mg levodopa9 F 58 5, 48 96, 36 3/40 3 5/2.5 Right 1400 mg levodopa; 2 mg cabergoline

10 M 68 5, 119 120, 72 17/43 4 5/3 Left 1400 mg levodopa11 F 41 13, 72 84, 18 15/42 4 2.5/1 Right 400 mg levodopa; 4.5 mg pramipexole12 M 59 8, 120 140, 82 15/51 5 3/1.5 Right 600 mg levodopa13 F 60 18, 128 152, 96 30/56 4 3/2 Right 800 mg levodopa; 6 mg ropinirole14 F 62 5, 160 168, 96 15/40 5 2.5/1.5 Right 600 mg levodopa15 M 47 18, 96 108, 36 10/29 5 2/1 Right 400 mg levodopa16 F 57 13, 120 132, 48 24/47 4 2.5/2 Right 600 mg levodopa17 F 54 5, 96 126, 36 31/56 4 4/2 Right 800 mg levodopa; 3 mg pramipexole18 M 57 5, 114 120, 72 17/56 5 4/1.5 Left 700 mg levodopa; 15 mg ropinirole19 F 39 8, 78 96, 12 32/56 5 4/2 Left 1100 mg levodopa; 3 mg pramipexole20 M 50 5, 96 108, 18 12/38 4 2/1 Left 400 mg levodopa21 M 68 5, 150 174, 48 19/56 4 3/1.5 Left 1000 mg levodopa; 6 mg pramipexole22 F 57 13, 112 136, 36 3/24 5 1.5/1 Left 400 mg levodopa; 4.5 mg pramipexole23 F 67 5, 168 180, 72 15/51 6 3/1.5 Left 800 mg levodopa; 15 mg ropinirole24 F 69 5, 156 168, 72 27/55 5 4/2 Right 800 levodopa; 1.5 mg pramipexole25 F 48 13, 78 108, 36 14/52 3 4/2 Left 600 mg levodopa; 15 mg ropinirole

a Variables expressed in months.

M. Amanzio et al. / Brain and Cognition 72 (2010) 337–346 339

assessed on the basis of Section III of the UPDRS. The score for eachitem on the scale ranged from 0 (absence of symptoms) to 4 (se-vere symptoms). Section IV-A concerning complications of therapywas also used to evaluate the duration of dyskinesias and relateddisability (for the importance of an accurate assessment of thesemotor complications, see Koller, Hutton, Tolosa, & Capilldeo,1999; Marras & Lang, 2003). The stage of disease was determinedusing the Hoehn and Yahr (1967) rating scale, range 1–5.

Neuropsychological assessment included MMSE to detect thepresence of a general cognitive deterioration; a modified version

of the Claridge Test (Claridge, 1967; Hume & Claridge, 1965) wasadministered for auditory selective attention and at the same timeto evaluate the impact of a psychomotor interference in a go-/no-go task condition. Executive functions were analyzed using themodified version of the Wisconsin Card Sorting Test, WCST (Laiac-ona, Inzaghi, De Tanti, & Capitani, 2000; Nelson, 1976) and the Pho-nemic Fluency Test (Novelli et al., 1986). Lastly, memory abilitieswere analyzed using subscales IV and VII of the Wechsler MemoryScale (WMS) (Wechsler, 1987). Neuropsychiatric assessment in-cluded the Hamilton Anxiety and Depression Scales (HAM-A and

340 M. Amanzio et al. / Brain and Cognition 72 (2010) 337–346

HAM-D, Hamilton, 1959, 1960) and the Brief Psychiatric RatingScale (BPRS) (Overall & Gorham, 1962). The neuropsychologistswho administered the tests and scales were blind to the aim ofthe study.

2.3. Assessment of awareness of movement disorders (see Appendix A)

Before undergoing the neuropsychological assessment all pa-tients were evaluated in terms of their awareness of movementdisorders using three brief scales: (1) the Global Awareness ofMovement (GAM) disorders scale, administered in the on and offstates (2A) the Dyskinesias Rating Scale and (2B) the Hypo-brady-kinesia Rating Scale.

(1) As far as the GAM was concerned, two simple rating scales,for global assessment of hypo/bradykinesia and dyskinesia aware-ness, were created by adapting them from the scale of anosognosiafor hemiplegia suggested by Bisiach, Vallar, Perani, Papagno, andBerti (1986). Each GAM consisted of 4 levels (0–3), ranging fromgood to no awareness of motor deficits. Scores were assigned bya neurologist experienced in movement disorders and based onthe degree of spontaneity with which subjects reported dyskine-sias in the on state versus hypokinesias/-bradykinesias in the offstate. Therefore, a high score corresponded to a high level ofunawareness of movement disorders in relation to symptoms ofdyskinesia and hypokinesia.

(2A and 2B) The severity of hypo-bradykinesias and dyskinesiaswas evaluated separately by the PD patients and the examiner. Pa-tients were requested to perform the following simple actions:write a short sentence, execute some verbal command. Scores wereassigned from 0 (total absence of Dyskinesia and/or Hypo-bradyki-nesia) to 3 (severe Dyskinesia and/or Hypo-bradykinesia).

Two awareness indexes were calculated by subtracting the pa-tients’ judgements from those of the neuropsychologists’ on hyperand hypo-bradykinesias awareness. The two variables were labeledas the Dyskinesias Subtracted Index (DS-I) and the Hypo-bradyki-nesias Subtracted Index (HS-I). Higher scores indicated a more se-vere impairment in terms of reduced awareness of movementdisorders.

2.4. Awareness of disabilities in activities of daily living and autonomylevel on the NUDS scale

The North University Disability Scale, NUDS (Canter et al., 1961)was used to evaluate the level of independence versus inability inbasic and instrumental activities of daily living, i.e. walking, gettingdressed, personal hygiene, eating and talking, considered sepa-rately in the on and off states.

The NUDS consists of 10 questions that were administered bothto our 25 patients and to one of their caregivers/significant other/close relatives, in order to compare the level of awareness of thedisability. Caregivers provided their evaluations in another room,separately from the patients, on the same day as the patient’sassessment. In analyzing the data we considered both patients’and caregivers’ evaluations of the patients’ disabilities. This scalewas chosen for its relatively good reliability and validity in mea-suring the impact of motor deficits in activities of daily living (Ra-maker, Marinus, Stiggelbout, & Van Hilten, 2002).

The awareness index was obtained by subtracting patients’evaluations of level of autonomy from caregivers’ evaluations ofthe patients’ level of disability, and labeled the NUDS SubtractedIndex (NUDS-I) for the on and off states.

2.5. Procedures

Parkinsonian subjects were assessed both the morning beforeand after their first daily dose in the off and on states; in particular

about 60–90 min after their morning dose of levodopa, during theperiod of maximum benefit from the medication. In the off statethey were assessed after about 12 h of therapeutic withdrawal.Participants were tested on two different consecutive days. Theywere randomly divided into two subgroups: the on subgroup with12 patients examined first in the on state and the following day inthe off state; and the off subgroup with 13 patients examined firstin the off state and the following day in the on state.

The neuropsychological assessment, in which patients had toperform on the Claridge, WCST modified version, Phonemic Flu-ency, subscales IV and VII of the WMS and MMSE were performedin about 1 h, whereas the complete evaluation session lastedapproximately 2 h.

2.6. Data analysis

Statistical analyses were performed using STATISTICA Softwarefor Windows (version 4.5 Stat Soft, Inc., 1993). Parametric statisticswere used since an initial exploration of the data set suggested anacceptable distribution (Skewness <1.00; Kurtosis <3.00). On theother hand, non-parametric statistics were used for the variablesassessing the awareness of movement disorders.

The sequence of the state effect (on–off versus off–on) was as-sessed using an ANOVA analysis with the group (on-first then-off; off-first then-on) as a between-subjects factor and the typeof neuropsychological measurement (i.e.: Claridge in-on versusClaridge in-off) as a within-subjects factor and post hoc tests. Onthe other hand, a series of Kruskal–Wallis tests were run to evalu-ate the effect of predominantly left versus predominantly right-side motor symptoms at the onset of disease on the awareness ofmovement disorders.

Finally, Spearman’s correlation coefficients were used to assessthe relationships between awareness measurements and othervariables. In this case we adopted a Bonferroni correction of the al-pha value to control for multiple tests.

3. Results

The caregivers obtained an MMSE score of 28.48 (±1.19) reveal-ing no difference with the MMSE scores of PD patients in the onstate.

3.1. Motor and cognitive data: on versus off states

Tables 1 and 2 show the demographic and clinical data of all PDsubjects and for each patient. Anti-parkinsonian medications wereexpressed as levodopa equivalent daily dosage (LEDD).

All PD patients had no significant difference between states ontheir MMSE scores. Parkinsonian patients had a low level of motorimpairment in the on state compared to the off state, assessed onthe basis of UPDRS (part III) data [ANOVA F(1, 48) = 161.991,p < 0.000001].

3.2. Awareness of movement disorders (see Tables 3 and 4)

Neurologists’ assessments of patients’ unawareness of move-ment disorders on the GAM were significantly higher for dyskineticthan for hypo-bradykinetic symptoms [Kruskal–WallisH(1, 50) = 23.550, p < 0.000001]. A comparison between patients’and neuropsychologists’ evaluations of awareness of dyskinesiasrevealed a significant difference in the on state [Kruskal–WallisH(1, 50) = 7.090, p = 0.0078], whereas no significant differencewas observed between patients’ and examiners’ evaluations ofawareness of hypokinesias in the off state [Kruskal–WallisH(1, 50) = 0.060, p = 0.807]. Hence, we found a significant differ-

Table 3Global Awareness of Movements (GAM) scale of dyskinesias and hypo-bradykinesiaon each patient in the on and off states. Higher scores indicate more severeimpairment in terms of reduced awareness of movement disorders.

Patient code GAM dyskinesias GAM hypo-bradykinesia

1 0 02 1 03 1 04 1 05 1 16 2 07 2 08 3 09 2 010 2 011 1 012 2 113 2 114 2 115 1 016 1 017 1 018 2 019 2 020 1 121 1 122 0 023 2 024 1 025 0 0M (±SD) 1.36 (0.76) 0.24 (0.43)

Table 4Awareness measures of each PD patient considering dyskinesias (DS-I) and hypo-bradykinesia (HS-I) motor symptoms. In most cases, considering DS-I, patientsevaluate their symptoms as less serious. DS-I: Dyskinesias Subtracted Index; HS-I:Hypo-Bradykinesia Subtracted Index.

Patientcode

On state Off state

DS-I Examiner Patient HS-I Examiner Patient

1 0 2 2 0 3 32 0 2 2 �1 1 23 1 1 0 1 3 24 1 3 2 0 2 25 1 3 2 0 2 26 0 1 1 0 2 27 0 2 2 �1 1 28 2 3 1 0 2 29 0 2 2 0 3 310 1 1 0 0 2 211 �1 2 3 0 3 312 1 3 2 0 3 313 1 2 1 0 3 314 1 3 2 0 2 215 0 3 3 0 3 316 1 2 1 0 3 317 1 2 1 0 2 218 1 3 2 0 3 319 1 3 2 �1 2 320 0 2 2 0 3 321 1 2 1 0 3 322 1 3 2 0 3 323 1 3 2 0 3 324 1 3 2 0 3 325 �1 1 2 0 2 2M (±SD) 0.60

(0.71)2.28(0.74)

1.68(0.75)

�0.08(0.40)

2.48(0.65)

2.56(0.51)

Table 5Awareness of disabilities in activities of daily living and autonomy level on the NUDSscale in the on and off states. NUDS-I: NUDS Subtracted Index.

Patientcode

On state Off state

NUDS-I Caregiver Patient NUDS-I Caregiver Patient

1 �3 13 10 �2 37 352 5 6 11 �1 25 243 �5 7 2 �8 12 44 �9 20 11 �4 37 335 �4 9 5 �5 16 116 �10 11 1 �3 26 237 �3 5 2 �3 40 378 2 12 14 �1 42 419 �1 6 5 �5 37 3210 �7 11 4 �13 35 2211 0 27 27 0 42 4212 �3 15 12 0 38 3813 �1 12 11 �11 31 2014 0 13 13 �13 32 1915 3 12 9 0 36 3616 �3 10 7 �1 32 3117 �5 7 2 0 28 2818 �1 23 22 �2 36 3419 �7 27 20 0 35 3520 0 15 15 1 30 3121 �6 20 14 0 33 3322 0 13 13 0 42 4223 �7 21 14 �25 44 1924 0 35 35 0 40 4025 �1 6 5 0 32 32M (±SD) �2.64

(3.73)14.24(7.71)

11.36(8.17)

�3.84(6.04)

33.52(7.74)

29.68(9.75)

The range of scores at NUDS scale is 0–50 points with higher scores indicating moresevere disabilities.

M. Amanzio et al. / Brain and Cognition 72 (2010) 337–346 341

ence between the two indexes of awareness of motor symptoms:Dyskinesias Subtracted Index (DS-I) and Hypo-bradykinesias Sub-tracted Index (HS-I), in the on and off states respectively, attestingthe presence of greater unawareness of dyskinesias [Kruskal–Wal-lis H(1, 50) = 14.997, p = 0.0001].

3.3. Awareness of disabilities in activities of daily living and autonomylevel on the NUDS scale (see Table 5)

A significant difference was observed when comparing disabil-ity assessment in the on and off states; more specifically, on theNUDS scale patients described themselves as less impaired in theon versus the off state [ANOVA F(1, 48) = 51.830, p < 0.000001].Likewise, caregivers described patients as less impaired in the onversus the off state [ANOVA F(1, 48) = 77.852, p < 0.000001]. Nodifferences were observed between the evaluation of the level ofimpairment by caregivers and patients in the on state [ANOVAF(1, 48) = 1.642, p = 0.206] and in the off state [ANOVAF(1, 48) = 2.378, p = 0.129].

3.4. Influence of sequence (on-first then-off versus off-first then-on) ondemographical and clinical variables

A one-way ANOVA was performed with sequence (on versusoff) as the between factor. No significant difference was observedbetween the two subgroups, in terms of age, education, durationof disease, Hoehn and Yahr stage in the on or off states, UPDRS mo-tor score in the on and off states, and dose of L-Dopa.

3.5. Influence of sequence (on-first then-off versus off-first then-on) onawareness assessment

The analysis revealed no significant difference between thegroups in terms of sequence and GAM in the on and off states[Kruskal–Wallis H(1, 25) = 0.0034769, p = 0.9530;H(1, 25) = 0.0121382, p = 0.9123 respectively] and sequence andDS-I [Kruskal–Wallis H(1, 25) = 0.0009692, p = 0.9752] and HS-I[Kruskal–Wallis H(1, 25) = 0.8814142, p = 0.3478]. Moreover, anal-ysis of NUDS-I showed no significant difference in terms of se-

342 M. Amanzio et al. / Brain and Cognition 72 (2010) 337–346

quence [ANOVA F(1, 23) = 2.7809, p = 0.1089] or in terms of NUDS-I in the on and off states [ANOVA F(1, 23) = 0.97999, p = 0.3325].The interaction between sequence and NUDS-I in the on versusoff states revealed no significant difference [ANOVAF(1, 23) = 0.1330, p = 0.7186].

3.6. Influence of sequence (on-first then-off versus off-first then-on) onneuropsychological and neuropsychiatric assessment

In particular, the neuropsychological test data were analyzedusing an ANOVA-mixed design with the ‘‘neuropsychological task”in the on and off states as the within-subjects factor and‘‘sequence” (on-first-then-off versus off-first-then-on) as the be-tween-subjects factor. For the Claridge Test the analysis showedno significant difference between groups in terms of sequence[F(1, 23) = 0.03, p = 0.87]; on the other hand the Claridge testshowed a significant difference in the off and on states[F(1, 23) = 5.18, p = 0.03]. We found no interaction between se-quence and this type of task performed in the on versus off states[F(1, 23) = 0.20, p = 0.66].

The analysis of executive functions measured using the WCSTmodified version and Phonemic Fluency Test showed no significantdifference as regards sequence [F(1, 23) = 0.007, p = 0.94;F(1, 23) = 0.33, p = 0.57, respectively] or the task performed in theon and off states [F(1, 23) = 1.75, p = 0.22; F(1, 23) = 0.79, p = 0.38,respectively]. The interaction between sequence and these typesof tasks performed in the on versus off states revealed no signifi-cant difference [F(1, 23) = 0.02, p = 0.90; F(1, 23) = 0.001, p = 0.97,respectively].

The analysis of memory measured on subscales IV and VII of theWMS showed no significant difference in term of sequence[F(1, 23) = 0.005, p = 0.94; F(1, 23) = 1.173, p = 0.69, respectively]and interaction between sequence and these types of tasks per-formed in the on versus off states [F(1, 23) = 1.03, p = 0.34;F(1, 23) = 2.44, p = 0.157, respectively]. Subscale IV of the WMSshowed no significant difference in terms of tasks performed inthe on and off states [F(1, 23) = 0.014, p = 0.91], whereas subscaleVII did [F(1, 23) = 8.61, p = 0.02].

The analysis of the neuropsychiatric assessment using the BPRS,HAM-A and HAM-D scales showed no significant difference interms of sequence [F(1, 23) = 0.36971, p = 0.549122;F(1, 23) = 1.22740, p = 0.279365; F(1, 23) = 0.6467, p = 0.429523,respectively]. The analysis revealed significant differences betweenthese three measures in the on and off states, in particular PD pa-tients scored higher in the off state compared to the on state[F(1, 23) = 42.90372, p = 0.000001; F(1, 23) = 55.98681,p < 0.000001; F(1, 23) = 168.3730, p < 0.000001, respectively]. Inparticular, for the BPRS, the higher scores in the off state weremainly due to items investigating anxiety and apathetic-depres-sive symptoms. The interaction between sequence and these threemeasures collected in the on versus off states did not reveal a sig-nificant difference [F(1, 23) = 0.72993, p = 0.401717;F(1, 23) = 0.02929, p = 0.865607; F(1, 23) = 1.9786, p = 0.172903respectively].

3.7. Influence of side of onset on awareness assessment

The analysis revealed no significant differences between thegroups in terms of side of onset and GAM in the on and off states[Kruskal–Wallis H(1, 25) = 0.9022219, p = 0.3422;H(1, 25) = 0.3499671, p = 0.5541 respectively] and side of onsetand DS-I [Kruskal–Wallis H(1, 25) = 0.1660, p = 0.6837] and HS-I[Kruskal–Wallis H(1, 25) = 0.0295188, p = 0.8636]. Moreover, theanalysis of NUDS-I showed no significant difference in terms ofside of onset [ANOVA F(1, 23) = 0.8797, p = 0.358] or in terms ofNUDS-I in the on and off states [ANOVA F(1, 23) = 0.9460,

p = 0.341]. The interaction between side of onset and NUDS-I inthe on versus off states revealed no significant difference [ANOVAF(1, 23) = 0.0027, p = 0.959].

3.8. Correlations between awareness of motor deficits and disability(NUDS) and the other variables

We used a Bonferroni correction of the alpha value to control forthe multiple test effect. For instance, given that we compared thethree measures of awareness for seven variables of interest (Cla-ridge and WCST modified version total scores, subscales IV andVII of the WMS, Phonemic Fluency Test, duration of disease andLEDD), the adjusted significance level was set at 0.05/7 = 0.007143.

In detail, in the on state, the global assessment of PD patients’levels of awareness of movement disorders (GAM) was signifi-cantly related to executive functions in terms of WCST modifiedversion total scores (R = �0.53, p = 0.006), suggesting that PD pa-tients with reduced awareness of motor deficits in the on state per-formed worse in this executive function task. In the same direction,we found a significant relationship between subtest IV of the WMSand the GAM scale (R = �0.61, p = 0.001). Finally, we observed apositive relationship between the GAM scale and LEDD (R = 0.55,p = 0.004).

On the contrary, in the off state we did not observe any relation-ship between global assessment of PD patients’ levels of awarenessof movement disorders (GAM) and the executive function vari-ables: WCST modified version total scores (R = �0.17, p = 0.63)and with the Phonemic Fluency Test (R = �0.01, p = 0.95). Nor didthe other variables appear to be related to GAM scores in the offstate.

Analyzing the relationship between DS-I and the other variableswe found no association with neuropsychological tests, but did ob-serve a relationship between DS-I and duration of disease (R = 0.55,p = 0.004), suggesting a more severe disease in patients with amore reduced awareness of their dyskinesia symptoms.

On the contrary, we did not observe significant relationships be-tween HS-I and other variables, and the NUDS subtracted Index(NUDS-I) in both the on and off states was not related with theother variables of interest.

4. Discussion

The primary aim of our study was to investigate the awarenessof different movement disorders in PD patients in the on versus theoff state. In our study, converging evidence of a lower awareness ofdyskinesias, compared to a better-preserved awareness of hypo-bradykinesia, resulted from different and independent evaluations.Significant differences were found in the comparison betweenawareness of dyskinesias in the on state and awareness of hypo-bradykinesias in the off state assessed by a neurologist usingGAM scales. In particular, the neurologist’s assessment showed re-duced awareness of dyskinesias in 22 out of 25 patients (see Ta-ble 3, patients with a score P1), but reduced awareness of hypo-bradykinetic movement disorders in only 6 out of 25 patients(see Table 3, patients with a score = 1). On the same line, levelsof movement disorder awareness in the on state appeared to be re-duced in our parkinsonian patients when analyzing the discrep-ancy observed between patients’ and neuropsychologists’assessments of the severity of dyskinesias (DS-I) and between pa-tients’ and neuropsychologists’ assessments of severity of hypo-bradykinesia (HS-I). We also observed a significant difference be-tween patients’ and neuropsychologists’ scores for dyskinesias inthe on state, whereas in the off state we did not observe a signifi-cant difference between patients’ and neuropsychologists’assessments.

M. Amanzio et al. / Brain and Cognition 72 (2010) 337–346 343

Low awareness of dyskinesias depends on a number of ele-ments, e.g. the complex influence of dopaminergic treatmenton cognitive, behavioral and executive functions. As far as cogni-tive status and awareness of movement disorders are concerned,previous studies in parkinsonian patients (Leritz et al., 2004;Seltzer et al., 2001) found a relationship between these, althoughthey did not analyze unawareness of motor problems differenti-ating the on and off states. In particular, Leritz et al. (2004) at-tested that unawareness may become more evident whengeneral cognition is more compromised, as in the later stagesof disease, when the cortical brain regions become more af-fected. In line with this observation, Seltzer et al. (2001), usingpatient-caregiver report discrepancies to assess awareness ofdeficits in a sample of AD and PD patients, showed that whenthe PD group was broken down into cognitively impaired andcognitively intact, the intact group demonstrated appropriateawareness across all domains; this study also suggested thatthe level of cognitive functioning is a critical mediating variablein awareness of deficits. This statement was not confirmed byour findings. In particular, in our cognitively intact PD patientswe only observed unawareness of movement disorders in theon state, attesting the importance of dopaminergic therapyinducing dyskinesias by acting on the basal ganglia and by stim-ulating mesocorticolimbic pathways. In analyzing cognitive func-tions in PD, Gotham, Brown, and Marsden (1988) observed apositive effect of levodopa on alternating fluency (we reachedthe same result with the Phonemic Fluency Test), but a negativeeffect on conditional associative learning. These differences couldbe explained by the fact that the dopaminergic effects on behav-ioral and cognitive functions depend on the baseline level of per-formance (Kimberg, D’Esposito, & Farah, 1997; Mattay et al.,2000; Mehta et al., 2000). In particular, if the circuit is depleted,dopaminergic treatment improves the function; on the contrary,a detrimental effect occurs in case of dopaminergic overstimula-tion of a non-depleted circuit. As far as prefrontal involvement inParkinson’s disease is concerned, dopaminergic treatment im-proves executive functions related to the cortical–subcorticalnetwork from the dorsolateral prefrontal cortex to the dorsalcaudate nucleus (for example, switching between two tasks asin the case of the Claridge test), which is dopamine depleted.On the contrary, the same dopaminergic treatment impairs func-tions connected to the orbitofrontal cortex–ventral striatal cir-cuitry, such as probabilistic reversal learning (Cools, Barker,Sahakian, & Robbins, 2001). A comparison between the on andoff states in our PD patients showed differences in clinical, neu-ropsychological and neuropsychiatric assessment. In detail, inthe on state, as expected, parkinsonian patients showed bettermotor performance than in the off phase of the disease (seeUPDRS-part III scores). Moreover, cognitive and neuropsychiatricassessment revealed differences between the off and on states.In particular, performance on the Claridge modified test was bet-ter in the on state, with fewer omissions and false alarms, sug-gesting a slight impairment in executive functions, in terms ofselective attention, in the off state. As also reported by otherauthors, the inability to perform several types of attention tasks(Pahwa & Koller, 1998; Raskin, Borod, & Tweedy, 1990; Ridenour& Dean, 1999) observed in PD patients appears to be predomi-nantly secondary to impaired inhibitory mechanisms (Zgaljardic,Borod, Foldi, & Mattis, 2003). Our patients in the off state alsoexhibited poor performance on a memory task of the VII sub-scale of the WMS, compared to the on state. Other authors haveattributed this phenomenon to frontal executive deficits (Tayloret al., 1986). In particular, difficulty in initiating and effectivelymaintaining search strategies, despite preserved encoding andrecognition, is reportedly mediated by the dorsolateral prefrontalcortex (Lichter, 2000), supporting the notion that the memory

impairments described in PD most likely reflect an executivedeficit possibly indicative of select frontostriatal circuitdisruption.

Concerning the complex influence of dopaminergic treatment onbehavioral, in the off state PD patients also showed higher levels ofanxiety, depressive and apathetic symptoms, as revealed by scoreson the BPRS, HAM-A and HAM-D scales. In particular, our PD subjectsin the on phase had a low level of depressive mood; in fact no pa-tients had been diagnosed as having major depression or dysthymiabased on DSM-IV criteria (1994). This is an important element to beconsidered when studying awareness, because depressed patientsmay view themselves negatively, and be biased toward reportingmore problems. Importantly, selected items of the psychiatric scalesconfirmed that most of our PD patients in the off state showed signsof apathetic behavior in terms of diminished goal-directed cognitionand behavior (lack of interest and reduced emotional responsive-ness). These neurobehavioral findings could be related to physicaldiscomfort due to parkinsonian symptoms; on the other hand, dopa-minergic treatment may also have a relief effect on apathetic symp-tomatology in the on state (Czernecki et al., 2002). Apatheticsyndrome is reported to be frequent in PD patients (Aarsland, Cum-mings, & Larsen, 2001; Aarsland et al., 1999; Aarsland, Litvan, & Lar-sen, 2001; Isella, Melzi, & Grimaldi, 2002; Pluck & Brown, 2002;Starkstein et al., 1992). In this particular case, basal ganglia dysfunc-tion due to nigro-striatal dopaminergic loss caused the capacity ofthe frontal cortex to select, initiate, maintain and shift a programof actions to be impaired (Levy & Dubois, 2006). Czernecki et al.(2002), using Starkstein’s apathy scale (Starkstein et al., 1992), dem-onstrated a significant difference in the severity of apathy betweenthe off and on states in fluctuating PD patients, suggesting that apa-thy is at least partly a dopamine–dependent syndrome. Both cogni-tive and behavioral findings observed in our parkinsonian patients inthe off state could be related to a slight dysfunction in the frontal–subcortical associative loops (Alexander, DeLong, & Strick, 1986).

Our hypothesis for explaining low awareness of dyskinesias isthat levodopa treatment may produce a detrimental effect on thefunction of the orbitofrontal and cingulated frontal–subcorticalloops; these projections appear to be critical in the awareness phe-nomenon (Leritz et al., 2004; Seltzer et al., 2001). Our parkinsonianpatients were selected for their intact cognitive status; however,low awareness of dyskinesias in the on state was found to be re-lated to poorer performance in the WCST and on a memory taskof subscale IV of the WMS. This finding supports the influence ofa reduced functionality in the anterior cingulated cortex, duringdopaminergic stimulation, even without excluding a role of thefrontostriatal circuits. A recent PET study on healthy subjects(Lumme, Aalto, Ilonen, Någren, & Hietala, 2007) actually showedthat errors in the WCST correlated with dopaminergic D2/D3 bind-ing in the right anterior cingulated cortex, suggesting a role of thisregion in executive functioning. Functional neuroimaging studieshave also clarified the roles of different prefrontal regions involvedin performing the WCST; in particular, Monchi, Petrides, Petre,Worsley, and Dagher (2001) demonstrated a role of the ventrolat-eral prefrontal cortex (BA 47/12) and anterior cingulate cortex (BA32) in terms of increased activity during the reception of negativefeedback from subjects on performance. Contextually, Monchi, Pet-rides, Mejia-Constain, and Strafella (2007) recently investigatedthe fMRI activation of the prefrontal cortex (PFC) in a group ofPD patients and healthy controls during the execution of the WCST.In the PD group, a significant decrease in activation was observedin those areas where activity, in healthy controls, was linked withthe striatum, namely the ventrolateral and the posterior PFC. Theauthors also observed a selective engagement of the dorsolateralPFC during the provision of feedback after each matching responseby the subjects (Ko, Monchi, Ptito, Petrides, & Strafella 2008); thisevidence is consistent with the hypothesis that dorsolateral PFC

1 The GAM scales were created by adapting them from the scale of Bisiach et al.(1986).

344 M. Amanzio et al. / Brain and Cognition 72 (2010) 337–346

activity is closely related to the monitoring of events in workingmemory.

As regards the influence of cognitive status on the awarenessphenomenon, the anterior cingulated cortex had a crucial role inthe control of action, such as attention-for-action-/target selection(Posner, Petersen, Fox, & Raichle, 1988), motor response selection(Paus, Petrides, Evans, & Meyer, 1993; Turken & Swick, 1999) anderror detection in performance monitoring (Gehring & Knight,2000; Luu, Flaisch, & Tucker, 2000). All these elements are impor-tant in awareness phenomena, as demonstrated in our study by thecorrelation between low awareness and executive functions. Lowawareness, observed in our PD patients, could be caused by an im-paired judgment capacity or metacognitive competence. In partic-ular, the unawareness of deficits related to movement disorders inthe on state observed in our patients may be seen as a function ofan inability to monitor one’s own cognitive abilities. One studyexamining speech-monitoring skills in AD and PD patients sup-ports the conclusion that both classes of patients are less able tomonitor their cognitive performance (McNamara, Obler, Au, Durso,& Albert, 1992). In particular, the failure to self-correct expressivespeech errors was thought to be related to attentional and frontaldysfunctions in these patients. In addition, Flowers and Robertson(1985) stated that PD patients seemed less able than controls tocheck their responses and inhibit errors during a cognitive task,suggesting they may be somewhat impaired in general responsemonitoring. In line with these observations, we found a relationbetween GAM and WCST and GAM and WMS IV scores on ‘‘med”only. This was not observed for Claridge and WMS VII, in whichprobably the general response monitor competence may have alower impact.

Judgement ability and metacognitive competence are likely in-volved in the self-assessment of activities of daily living; the com-parison between NUDS-I in the on state and NUDS-I in the off statedid not show any difference; this finding suggests that levodopatreatment does not affect a global cognitive system of awareness,but only has a specific effect on a system that monitors motorbehavior in terms of movement disorders. In the comparison be-tween patients’ and caregivers’ evaluations on the NUDS scale,caregivers did not judge patients to be more impaired in eitherthe off or on states compared to the patients’ own evaluations oftheir disabilities. These findings seem to exclude the hypothesisof a role of pessimistic evaluation of parkinsonian patients by theircaregivers (Carter et al., 1998; Martínez-Martín et al., 2004).

The main interpretative problem related to our topic was thedifficulty of disentangling the role of levodopa treatment (on stateversus off state) and the role of movement disorder (hyperkinesiaversus hypokinesia) in low awareness of dyskinesias. Low aware-ness of hyperkinesias has been found in other neurological dis-eases, for example in Huntington’s disease (Deckel & Morrison,1996; Ho, Robbins, & Barker, 2006; Snowden, Craufurd, Griffiths,& Neary, 1998; Vitale et al., 2001), in tardive dyskinesia (Cohen &Cohen, 1993) and in patients with lesions of the basal ganglia (Laz-zarino & Nicolai, 1991). These studies may support a crucial rela-tionship between dyskinesias and low awareness observed in ourstudy. Dyskinesias are usually absent in parkinsonian patients inthe off state; however, if we consider the hyperkinetic domain ofmovement disorders, resting tremor can be found in the off state.It is important to underline that we did not investigate awarenessof tremor in our patients because in many cases this symptom wasabsent or only slight and had no important influence on actions.Even in those few cases, patients with tremor did not showunawareness of this hyperkinetic symptom in the off state.

In PD patients the progression of dopaminergic degeneration,also in situations complicated by motor fluctuations, as in our par-kinsonian group, could involve the dorsal part of the striatum morethan the ventral striatum (Agid et al., 1993; Kish et al., 1988). Levo-

dopa treatment limits motor impairment, associated with the dor-sal striatum; therefore, high dosages of levodopa improveparkinsonian motor symptoms but could have a detrimental effecton the ventral system and account for the low awareness of dyski-nesias in the on state; the relationship between low awareness ofdyskinesias (GAM-scale) in the on state and LEDD supports thishypothesis. In line with this interpretation, we observed relation-ships between the discrepancy between patients’ and neuropsy-chologists’ assessments of severity of dyskinesias (DS-I) andduration of disease. Both duration and severity of disease may beassociated with frontal lobe impairment (especially of the ventro-lateral prefrontal cortex) as shown in HD subjects (Folstein, Fol-stein, & Brandt, 1990).

It is also important to underline the relationship between themesocorticolimbic dopaminergic system and reward mechanismsin PD patients. This appears to be an important point to be consid-ered in this area of research as some patients with particular neu-ropsychiatric disorders such as Hedonistic HomeostaticDysregulation (HHD) develop severe but surprisingly well-toler-ated drug-induced dyskinesias, which do not particularly botherthem. Since HHD is not a common phenomenon (Giovannoni,O’Sullivan, Turner, Manson, & Lees, 2000), and this psychiatric dis-order was not represented in our PD population, we can excludethat this type of dysfunction in the mesocorticolimbic dopaminer-gic system related to reward and addiction represents a biasingfactor in our results. However, we are aware of the importance ofincluding specific tests in future studies to directly explore theactivity of the mesocorticolimbic network, i.e. probabilistic learn-ing (Cools et al., 2001) and the prospect of a reward.

In conclusion, our findings demonstrate that unawareness ofmovement disorders may also occur in cognitively unaffected PDpatients. This element has important implications for managingthis particular class of pathology. In particular, low awareness ofdyskinesias in the on state appears to be related to metacognitivedeficits in the self-monitoring system. Finally, in our study we didnot find the side of onset of disease to be salient in predictingunawareness of movement disorders as shown by Leritz et al.(2004). A possible explanation for this difference is the selectionof PD patients. In particular, we selected a sample with a more se-vere degree of the disease, in which the degeneration asymmetriesare probably attenuated.

Acknowledgments

The authors thank the three anonymous reviewers for theircontribution to the improvement of the manuscript. This workwas supported by grants from Fondazione CRT, Progetto Alfieri.

Appendix A

A.1. The Global Awareness of Movement (GAM) Disorders scale1

(0) When questioned by the examiner, about his/her state ofhealth, the patient spontaneously reported the presence ofinvoluntary movements (in the on state if present) or motorhindrance (in the off state if present).

(1) The patient reported the presence of involuntary move-ments (in-on) and motor hindrance (in-off), but only afteran explicit request by the examiner concerning those motordisturbances.

M. Amanzio et al. / Brain and Cognition 72 (2010) 337–346 345

(2) The patient only admitted the presence of involuntarymovements and motor hindrance after focusing attentionon his/her evident dyskinesias on any part of the body andafter a request to perform fine movements (i.e. a prona-tion/-supination hand task), which was done with evidentdifficulty or slowness (not due to his/her dyskinesias).

(3) The patient did not admit the presence of his/her evidentmotor disturbances, even after the above mentioneddemonstrations.

A.2. The patient was requested to perform the following three simpleactions

– Write down a sentence: ‘‘Oggi e’ una bella giornata di prima-vera” (Today is a lovely spring day).

– Hold a half-full glass in his/her hands, bring it up to the mouthand put it down again (patient is sitting at the table).

– Stand up from a chair, walk for two meters, go back to the chairand sit down again.

The actions were rated using the following scales of dyskinesias[2A] and hypo-bradykinesias [2B]

2A. Dyskinesias Rating Scale

(0) Absence of dyskinesias (unusual and involuntarymovements).

(1) Slight dyskinesias (slightly visible).(2) Moderate dyskinesias (clearly evident but without signifi-

cant impact on the result of the execution of the actions,except in particularly fine tasks).

(3) Severe dyskinesias (clearly evident and with a negativeimpact in the result of the execution of the actions).

2B. Hypo-bradykinesias rating scale

(0) Absence of hypo-bradykinesias (difficulty or slowness ofmovements).

(1) Slight hypo-bradykinesias (slightly evident, only during theexecution of fine movements, such as writing).

(2) Moderate hypo-bradykinesias (clearly evident but withoutimpact on the execution of actions except in finemovements).

(3) Severe hypo-bradykinesias (clearly evident and impactingon the execution of actions).

References

Aarsland, D., Cummings, J. L., & Larsen, J. P. (2001). Neuropsychiatric differencesbetween Parkinson’s disease with dementia and Alzheimer’s disease.International Journal of Geriatric Psychiatry, 16, 184–191.

Aarsland, D., Larsen, J. P., Lim, N. G., Janvin, C., Karlsen, K., Tandberg, E., et al. (1999).Range of neuropsychiatric disturbances in patients with Parkinson’s disease.Journal of Neurology, Neurosurgery and Psychiatry, 67, 492–496.

Aarsland, D., Litvan, I., & Larsen, J. P. (2001). Neuropsychiatric symptoms of patientswith progressive supranuclear palsy and Parkinson’s disease. The Journal ofNeuropsychiatry and Clinical Neuroscience, 13, 42–49.

Agid, Y., Ruberg, M., Javoy-Agid, F., Hirsch, E., Raisman-Vozari, R., Vyas, S., Faucheux,B., et al. (1993). Are dopaminergic neurons selectively vulnerable to Parkinson’sdisease? Advances in Neurology, 60, 148–164.

Alexander, G. E., DeLong, M. R., & Strick, P. L. (1986). Parallel organization offunctionally segregated circuits linking basal ganglia and cortex. Annual Reviewof Neuroscience, 9, 357–381.

Amanzio, M., & Torta, D. M. E. (2009). Unawareness of deficits in Alzheimer’s diseasethrough a biopsychosocial perspective. In Clayton B. Larson (Ed.), Metacognition:New research developments (pp. 239–253). Hauppauge, New York: Nova SciencePublishers.

American Psychiatric Association (1994). Diagnostic and statistical manual of mentaldisorders (4th ed.). Washington, DC: APA.

Baddeley, A. (1986). Working memory. New York: Oxford University Press.

Bisiach, E., & Geminiani, G. (1991). Anosognosia related to hemiplegia andhemianopia. In G. Prigatano & D. L. Schacter (Eds.), Awareness of deficit afterbrain injury (pp. 17–39). New York: Oxford University Press.

Bisiach, E., Vallar, G., Perani, D., Papagno, C., & Berti, A. (1986). Unawareness ofdisease following lesions of the right hemisphere: Anosognosia for hemiplegiaand anosognosia for hemianopsia. Neuropsychologia, 24, 471–482.

Canter, G. J., De Latorre, R., & Mier, M. (1961). A method for evaluating disability inpatients with Parkinson’s disease. Journal of Nervous and Mental Disease, 133,143–152.

Carter, J. H., Stewart, B. J., Archbold, P. G., Inoue, I., Jaglin, J., Lannon, M., et al. (1998).Living with a person who has Parkinson’s disease: The spouse’s perspective bystage of disease. Parkinson’s study group. Movement Disorders, 13(1), 20–28.

Claridge, G. S. (1967). Personality and arousal. A psychophysiological study ofpsychiatric disorders. Pergamon Press.

Cohen, H., & Cohen, D. (1993). What may be gained from neuropsychologicalinvestigations of tardive dyskinesia? Brain Cognition, 23(1), 1–7.

Cools, R., Barker, R. A., Sahakian, B. J., & Robbins, T. W. (2001). Enhanced or impairedcognitive function in Parkinson’s disease as a function of dopaminergicmedication and task demands. Cerebral Cortex, 11(12), 1136–1143.

Czernecki, V., Pillon, B., Houeto, J. L., Pochon, J. B., Levy, R., & Dubois, B. (2002).Motivation, reward and Parkinson’s disease: Influence of dopatherapy.Neuropsychologia, 2257–2267.

Deckel, A. W., & Morrison, D. (1996). Evidence of a neurologically based ‘‘denial ofillness” in patients with Huntington’s disease. Archives of ClinicalNeuropsychology, 11(4), 295–302.

Fahn, S., & Elton, R. L. (1987). Unified Parkinson’s disease rating scale. In C. D. Fahn,M. Marsden, M. Goldstein, & D. B. Calne (Eds.), Recent developments inParkinson’s disease. Florham, New York: McMillan Healthcare Information.

Flowers, F. A., & Robertson, C. (1985). The effect of Parkinson’s disease on the abilityto maintain a mental set. Journal of Neurology, Neurosurgery and Psychiatry, 48,517–529.

Folstein, M. F., Folstein, S. E., & Brandt, J. (1990). Huntington’s disease. In J. L.Cummings (Ed.), Subcortical dementia. New York: Oxford.

Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). Mini-mental state. Journal ofPsychiatric Research, 12, 189–198.

Gehring, W. J., & Knight, R. T. (2000). Prefrontal–cingulate interactions in actionmonitoring. Nature Neuroscience, 3(5), 516–520.

Giovannoni, G., O’Sullivan, J. D., Turner, K., Manson, A. J., & Lees, A. J. (2000).Hedonistic homeostatic dysregulation in patients with Parkinson’s disease ondopamine replacement therapies. Journal of Neurology, Neurosurgery andPsychiatry, 68, 423–428.

Godefroy, O., Rousseaux, M., Pruvo, J. P., Cabaret, M., & Leys, D. (1994).Neuropsychological changes related to unilateral lenticulostriate infarcts.Journal of Neurology, Neurosurgery and Psychiatry, 57(4), 480–485.

Gotham, A. M., Brown, R. G., & Marsden, C. D. (1988). ‘‘Frontal” cognitive functionin patients with Parkinson’s disease ‘‘on” and ‘‘off” levodopa. Brain, 111,299–321.

Hamilton, M. A. (1959). The assessment of anxiety states by rating. British Journal ofMedical Psychology, 32, 50–55.

Hamilton, M. A. (1960). A rating scale for depression. Journal of Neurology,Neurosurgery and Psychiatry, 23, 56–62.

Healton, E. B., Navarro, C., Bressman, S., & Brust, J. C. (1982). Subcortical neglect.Neurology, 32(7), 776–778.

Ho, A. K., Robbins, A. O., & Barker, R. A. (2006). Huntington’s disease patients haveselective problems with insight. Movement Disorders, 21(3), 385–389.

Hoehn, M. M., & Yahr, M. D. (1967). Parkinsonism: Onset, progression and mortality.Neurology, 17, 427–442.

Hughes, A. J., Daniel, S. E., Kilford, L., & Lees, A. J. (1992). The accuracy of clinicaldiagnosis of idiopathic Parkinson’s disease: A clinicopathological study. Journalof Neurology, Neurosurgery and Psychiatry, 55, 181–184.

Hume, W. I., & Claridge, G. S. (1965). A comparison of two measures of ‘‘arousal” innormal subjects. Life Science, 4, 545–553.

Isella, V., Melzi, P., & Grimaldi, M. (2002). Clinical, neuropsychological, andmorphometric correlates of apathy in Parkinson’s disease. MovementDisorders, 17, 366–371.

Jacome, D. E. (1986). Subcortical prosopagnosia and anosognosia. American Journalof Medical Science, 292(6), 386–388.

Jellinger, K. (1987). Overview of morphological changes in Parkinson’s disease.Advances in Neurology, 45, 1–18.

Kimberg, D. Y., D’Esposito, M., & Farah, M. J. (1997). Effects of bromocriptine onhuman subjects depend on working memory capacity. Neuroreport, 8(16),3581–3585.

Kish, S. J., Shannak, K., & Hornykiewicz, O. (1988). Uneven pattern of dopamineloss in the striatum of patients with idiopathic Parkinson’s disease. Patho-physiologic and clinical implications.. New England Journal of Medicine, 318(14),876–880.

Ko, J. H., Monchi, O., Ptito, A., Petrides, M., & Strafella, A. P. (2008). Repetitivetranscranial magnetic stimulation of dorsolateral prefrontal cortex affectsperformance of the Wisconsin card sorting task during provision of feedback.International Journal of Biomedical Imaging, 2008, 143238.

Koller, W. C., Hutton, J. T., Tolosa, E., & Capilldeo, R. (1999). Immediate-release andcontrolled-release carbidopa/levodopa in PD: A 5 years randomized multicenterstudy. Carbidopa/Levodopa Study Group. Neurology, 53, 1012–1019.

Laiacona, M., Inzaghi, M. G., De Tanti, A., & Capitani, E. (2000). Wisconsin CardSorting test: A new global score, with Italian norms, and its relationship withthe Weigl sorting test. Neurological Science, 21, 279–291.

346 M. Amanzio et al. / Brain and Cognition 72 (2010) 337–346

Lazzarino, L. G., & Nicolai, A. (1991). Hemichorea-hemiballism and anosognosiafollowing a contralateral infarction of the caudate nucleus and anterior limb ofthe internal capsule. Rivista di Neurologia, 61(1), 9–11.

Leritz, E., Loftis, C., Crucian, G., Friedman, W., & Bowers, D. (2004). Self-Awareness ofdeficits in Parkinson disease. The Clinical Neuropsychologist, 18, 352–361.

Levy, R., & Dubois, B. (2006). Apathy and the functional anatomy of the prefrontalcortex-basal ganglia circuits. Cerebral Cortex, 16, 916–928.

Lichter, D. G. (2000). Movement disorders and frontal-subcortical circuits. In D. G.Lichter & J. L. Cummings (Eds.), Frontal-subcortical circuits in psychiatric andneurological disorders (pp. 260–313). New York: The Guilford Press.

Lopez, O. L., Becker, J. T., Somsak, D., Dew, M. A., & DeKosky, S. T. (1994). Awarenessof cognitive deficits and anosognosia in probable Alzheimer’s disease. EuropeanJournal of Neurology, 34, 277–282.

Lumme, V., Aalto, S., Ilonen, T., Någren, K., & Hietala, J. (2007). Dopamine D2/D3receptor binding in the anterior cingulate cortex and executive functioning.Psychiatry Research, 156(1), 69–74.

Luu, P., Flaisch, T., & Tucker, D. M. (2000). Medial frontal cortex in actionmonitoring. Journal of Neuroscience, 20(1), 464–469.

Marras, C., & Lang, E. (2003). Measuring motor complications in clinical trials forearly Parkinson’s disease. Journal of Neurology, Neurosurgery and Psychiatry, 74,143–146.

Martínez-Martín, P., Benito-León, J., Alonso, F., Catalán, M. J., Pondal, M., &Zamarbide, I. (2004). Health-related quality of life evaluation by proxy inParkinson’s disease: Approach using PDQ-8 and EuroQoL-5D. MovementDisorders, 19(3), 312–318.

Mattay, V. S., Callicott, J. H., Bertolino, A., Heaton, I., Frank, J. A., Coppola, R., et al.(2000). Effects of dextroamphetamine on cognitive performance and corticalactivation. NeuroImage, 12(3), 268–275.

McGlynn, S. M., & Kaszniak, A. W. (1991). When metacognition fails: Impairedawareness of deficit in Alzheimer’s disease. Journal of Cognitive Neuroscience, 3,183–189.

McGlynn, S. M., & Schacter, D. L. (1989). Unawareness of deficit in neuropsy-chological syndrome. Journal of Clinical and Experimental Neuropsychology, 11,143–205.

McNamara, P., Obler, L. K., Au, R., Durso, R., & Albert, M. L. (1992). Speechmonitoring skills in Alzheimer’s disease, Parkinson’s disease, and normal aging.Brain and Language, 42, 38–51.

Mehta, M. A., Owen, A. M., Sahakian, B. J., Mavaddat, N., Pickard, J. D., & Robbins, T.W. (2000). Methylphenidate enhances working memory by modulating discretefrontal and parietal lobe regions in the human brain. Journal of Neuroscience,20(6), 240–265.

Monchi, O., Petrides, M., Mejia-Constain, B., & Strafella, A. P. (2007). Cortical activityin Parkinson disease during executive processing depend on striatalinvolvement. Brain, 130, 233–244.

Monchi, O., Petrides, M., Petre, V., Worsley, K., & Dagher, A. (2001). Wisconsin CardSorting revisited: Distinct neural circuits participating in different stages of thetask identified by event-related functional magnetic resonance imaging.Neuroscience, 21(19), 7733–7741.

Myslobodsky, M. S. (1986). Anosognosia in tardive dyskinesia: ‘‘Tardive dysmentia”or ‘‘tardive dementia?”. Schizophrenia Bulletin, 12(1), 1–6.

Nelson, H. E. (1976). A modified Card Sorting Test sensitive to frontal lobe defect.Cortex, 12, 13–24.

Novelli, G., Papagno, C., Capitani, E., Laiacona, M., Vallar, G., & Cappa, S. F. (1986).Three clinical tests for the assessment of lexical retrieval and production. Normfrom 320 normal subjects. Archivio di Psicologia, Neurologia e Psichiatria, 47,477–506.

Overall, J. E., & Gorham, D. R. (1962). The brief psychiatric rating scale. PsychologicalReports, 10, 799–812.

Owen, A. M., James, M., Leigh, P. N., Summers, B. A., Marsden, C. D., Quinn, N. P.,Lange, K. W., et al. (1992). Fronto-striatal cognitive deficits at different stages ofParkinson’s disease. Brain, 115(6), 1727–1751.

Pahwa, R., & Koller, W. C. (1998). Advances in the treatment of Parkinson’s disease.Drugs Today (Barc), 34(2), 95–105.

Paus, T., Petrides, M., Evans, A. C., & Meyer, E. (1993). Role of the human anteriorcingulate cortex in the control of oculomotor, manual, and speech responses: Apositron emission tomography study. Journal of Neurophysiology, 70(2),453–569.

Pezzella, F. R., Di Rezze, M., Chianese, M., Fabbrini, G., Vanacore, N., Colosimo,C., & Meco, G. (2003). Hedonistic homeostatic dysregulation in Parkinson’sdisease: A short screening questionnaire. Neurological Science, 24, 205–206.

Pluck, G. C., & Brown, R. G. (2002). Apathy in Parkinson’s disease. Journal ofNeurology, Neurosurgery and Psychiatry, 73, 636–642.

Posner, M. I., Petersen, S. E., Fox, P. T., & Raichle, M. E. (1988). Localizationof cognitive operations in the human brain. Science, 240(4859), 1627–1631.

Ramaker, C., Marinus, J., Stiggelbout, A. M., & Van Hilten, B. J. (2002). Systematicevaluation of rating scales for impairment and disability in Parkinson’s disease.Movement Disorders, 17(5), 867–876.

Raskin, S. A., Borod, J. C., & Tweedy, J. (1990). Neuropsychological aspects ofParkinson’s disease. Neuropsychological Review, 1(3), 185–221.

Ridenour, T. A., & Dean, T. S. (1999). Parkinson’s disease and neuropsychologicalassessment. International Journal of Neuroscience, 99(1–4), 1–18.

Seltzer, B., Vasterling, J. J., Mathias, B. A., & Brennan, A. (2001). Clinical andneuropsychological correlates of impaired awareness of deficits in Alzheimerdisease and Parkinson disease: A comparative study. Neuropsychiatry,Neuropsychology, and Behavioural Neurology, 14, 122–129.

Shenker, J. I., Wylie, S. A., Fuchs, K., Manning, C. A., & Heilman, K. M. (2004). On-lineanosognosia: Unawareness for chorea in real time but not on videotape delay.Neurology, 63(1), 159–160.

Snowden, J. S., Craufurd, D., Griffiths, H. L., & Neary, D. (1998). Awareness ofinvoluntary movements in Huntington disease. Archives of Neurology, 55(6),801–805.

Starkstein, S. E., Mayberg, H. S., Preziosi, T. J., Andrezejewski, P., Leiguarda, R., &Robinson, R. G. (1992). Reliability, validity and clinical correlates of apathy inParkinson’s disease. The Journal of Neuropsychiatry and Clinical Neurosciences, 4,134–139.

Starkstein, S. E., Sabe, L., Petracca, G., Chemerinski, E., Kuzis, G., Morello, M., &Leiguarda, R. (1996). Neuropsychological and psychiatric differences betweenAlzheimer’s disease and Parkinson’s disease with dementia. Journal ofNeurology, Neurosurgery and Psychiatry, 61, 381–387.

Taylor, A. E., Saint-Cyr, J. A., & Lang, A. E. (1986). Frontal lobe dysfunction inParkinson’s disease. The cortical focus of neostriatal outflow. Brain, 109(5),845–883.

Turken, A. U., & Swick, D. (1999). Response selection in the human anteriorcingulate cortex. Nature Neuroscience, 2(10), 920–924.

Vitale, C., Pellecchia, M. T., Grossi, D., Fragassi, N., Cuomo, T., Di Maio, L., & Barone, P.(2001). Unawareness of dyskinesias in Parkinson’s and Huntington’s diseases.Neurological Science, 22, 105–106.

Wechsler, D. (1987). Wechsler Memory Scale—Revised manual. San Antonio: ThePsychological Corporation.

Weinstein, E. A., & Kahn, R. L. (1950). The syndrome of anosognosia. AMA Archives ofNeurology and Psychiatry, 64(6), 772–791.

Zgaljardic, D. J., Borod, J. C., Foldi, N. S., & Mattis, P. (2003). A review of the cognitiveand behavioural sequelae of Parkinson’s disease: Relationship to frontostriatalcircuitry. Cognitive and Behavioural Neurology, 16, 193–210.