Movement Disorders

Vol. 18, No. 2, 2003, pp. 150 –156

© 2002 Movement Disorder Society

Deficit of Verb Generation in Nondemented Patients with

Parkinson’s Disease

Patrice Péran, MSc,1 Olivier Rascol, MD, PhD,1 Jean-François Démonet, MD, PhD,1
Pierre Celsis, MD, PhD,1 Jean-Luc Nespoulous, PhD,2 Bruno Dubois, MD, PhD,3
and Dominique Cardebat, PhD1*
1
Institut National de la Santé et de la Recherche Médicale U 455, Centre Hospitalier Universitaire Purpan, Toulouse, France
2
Laboratoire Jacques Lordat, Toulouse, France
3
Institut National de la Santé et de la Recherche Médicale E007, Hôpital de la Salpêtrière, Paris, France

Abstract: Clinical and neuroimaging studies have shown that
verb processing suggests a preferential participation of a pre-
frontal network, which is dysfunctional in Parkinson’s disease
(PD). To assess a verb processing deficit in PD, we compared
noun- and verb-generation tasks for 34 nondemented PD pa-
tients (according to the Dementia Rating Scale) with 34
matched normal subjects, using two intracategory tasks (noun/
noun and verb/verb generation) and two intercategory tasks
(noun/verb and verb/noun generation). PD patients were sig-
nificantly impaired in the two tasks involving verb production,
i.e., verb/verb and noun/verb generation, whereas their perfor-
mance was similar to those of controls in the two tasks requir-
ing noun production. For the two impaired tasks, we assessed
1) the influence of lexical competition that corresponds to the
presence of several candidate words for a given stimulus; 2) the
influence of slight cognitive dysfunction; and 3) the influence
of motor deficit. Significant correlations were found between
DRS scores and performance on the noun/verb task, and no
significant correlations were found between lexical competition
or motor deficit and performance. The specific deficit for verb
production in PD patients is discussed in relation to deficits
affecting either action or grammatical representations. © 2002
Movement Disorder Society
Key words: Parkinson; language; word generation; verb

Several cognitive deficits can be observed in nonde-
mented patients with Parkinson’s disease (PD), even at
the early stages of the disease, that mainly reflect diffi-
culties with the initiation and execution of internally
guided information processing strategies. These disor-
ders, generally interpreted as the witness of the striato-
frontal loop pathophysiology1 affect visuospatial pro-
cessing, working and long-term memory, and executive
functions. Nevertheless, specific language disturbances
do not seem to form part of the PD cognitive pattern,
although several authors have intended to evidence def-
icits in semantic or letter verbal fluency that are consid-
ered as “frontal” language tests. Until now, verbal flu-
*Correspondence to: Dominique Cardebat, INSERM U 455, CHU
Purpan, 31059 Toulouse Cedex 3, France.
E-mail: cardebat@toulouse.inserm.fr
Received 4 February 2002; Revised 29 April 2002; Accepted 3 July
2002
ency deficit is not found systematically in PD patients
free of significant cognitive decline; indeed, contradic-
tory results have been reported, with studies showing no
lexical or semantic fluency deficits (see Hanley et al.,
Bayles et al., Troster et al., and Piatt et al.2–5), whereas
other authors identified deficits on semantic fluency but
not on lexical fluency (see Auriacombe et al.6). Recently,
and besides classic lexical and semantic fluency, Piatt
and colleagues5 explored action-verbal fluency in PD
demented and nondemented patients. They did not show
any particular deficit in action-verbal fluency in nonde-
mented PD patients compared to controls, but action
fluency appeared particularly sensitive to PD-associated
dementia. The absence of results for nondemented PD
patients can be explained by the difficulty of finding an
appropriate cue to induce verb production in verbal
fluency.
Indeed, dissociation in lexical categories may also be
relevant to explore language-related dysfunction of the
frontal cortex, because clinical evidence7–11 shows dou-

150
 DEFICIT IN VERB PROCESSING IN PD
TABLE 1. Demographic and clinical data
Parameter
Age (yr)
Education (ⱖ9 yr/⬍9 yr)
Gender (M/F)
MMSE
DRS total (/144)
Disease duration (yr)
UPDRS motor score
Levodopa (range, mg/day)
Dopamine agonists (range, mg/day)
Other antiparkinsonian drugs (range, mg/day)
Values are expressed as mean ⫾ SD.
MMSE, Mini-Mental State Examination; DRS, Dementia Rating Scale; UPDRS, Unified Parkinson’s Disease Rating Scale; PD, Parkinson’s
disease.
ble dissociations with a relationship between object-
naming deficit and lesions centred on the left temporal
lobe and, alternatively, action-naming deficit and large
lesions involving the left frontal cortex. These observa-
tions led Damasio and Tranel9 to formulate the hypoth-
esis that noun retrieval is preferentially mediated by the
left temporal lobe, whereas verb retrieval is mediated by
a large cerebral network including the left prefrontal
area. Although these clinical findings were only partly
replicated by neuroimaging studies in normal subjects
(for an exception, see Tyler et al.12), several neuroimag-
ing results showed the implication of left frontal regions,
and particularly Brodmann areas 44, 45, and 47 for verb
processing along with many other regions of the lan-
guage network.13–19
To explore the influence of the “frontal” dysfunction
on language production in nondemented PD patients, we
decided to use word-generation tasks that require a se-
mantics- and grammar-driven selection of a single word
over a limited time period. This kind of task alleviates
both the working memory load and the planification
process whose influence might account for the deficits
observed in PD in verbal fluency tasks.
We investigated word-generation processing in two
groups (34 PD nondemented patients and 34 matched
control subjects) by comparing verb- and noun-gener-
ation performance. If verb processing suggests a pref-
erential participation of a prefrontal network, a dis-
proportionate impairment in verb processing tasks
compared to noun processing tasks can be expected in
PD patients, due to the frontal pathophysiology of the
disease.
151
Control subjects (N ⫽ 34) PD subjects (N ⫽ 34)
60.3 ⫾ 10.4 63.3 ⫾ 8.9
16/18 16/18
25/9 25/9
28.8 ⫾ .8 27.8 ⫾ 1.6
— 140.6 ⫾ 3.7
— 8.5 ⫾ 6.6
— 18.9 ⫾ 8.5
— 1153.9 ⫾ 845.1 (n ⫽ 29; 62.5–4375)
— Bromocriptine (n ⫽ 10; 15–60)
Amantadine (n ⫽ 7; 100–900)
Ropinirol (n ⫽ 9; 9–20)
Lisuride (n ⫽ 2; 2.7 and 6.5)
— Entacapone (n ⫽ 5; 200–1,000)
Selegiline (n ⫽ 2; 5 and 10)
SUBJECTS AND METHODS
Subjects
Thirty-four right-handed French PD patients were in-
cluded in the study. Parkinson’s disease was diagnosed
by a staff neurologist on the basis of akinesia associated
with one of the other two cardinal signs (tremor, rigidity)
and responsiveness to levodopa therapy. All patients
fulfilled the UK Parkinson’s Disease Brain Bank criteria
for the diagnosis of idiopathic Parkinson’s disease.20
None of the patients had a history of neurological or
psychiatric disease other than PD. All patients were on
antiparkinsonian medication (29 on levodopa, 5 on do-
paminergic agonist) at the time of testing, and the pa-
tients were not taking medication associated with seda-
tion such as anticholinergics. The motor disability of
patients was evaluated using the motor part of the Uni-
fied Parkinson’s Disease Rating Scale (UPDRS).21 Pa-
tients who had end-of-dose motor fluctuations were al-
ways tested in the on phase. Cognitive status was
assessed using Mini-Mental State Examination22
(MMSE; mean, 27.8 ⫾ 1.6; range, 25–30). Patients who
scored less than 25 were excluded to avoid inclusion of
demented PD patients. Moreover, a global assessment of
cognitive functions was obtained using the Dementia
Rating Scale23 (DRS; mean, 140.6 ⫾ 3.7; range, 128 –
144; 2 patients scored ⬍134).
A control group of 34 right-handed French subjects,
closely matched to PD patients for age, sex, and educa-
tion, was included in the study. The MMSE mean score
was 28.8 ⫾ 0.8. Demographic and clinical data of the
two groups are reported in Table 1.
Movement Disorders, Vol. 18, No. 2, 2003
152 P. PÉRAN ET AL.
Materials
Word generation was assessed with 40 concrete nouns
and 40 action verbs matched for lexical frequency24 and
length (di- and trisyllabic). The protocol25 consisted in
two intracategory tasks: noun-to-noun (NN) and verb-to-
verb (VV) generation, in which subjects were instructed
to produce a semantically related noun or verb when
listening to a noun or a verb; and two intercategory tasks:
noun-to-verb (NV) and verb-to-noun (VN) generation, in
which subjects were presented with either a noun or a
verb and instructed to produce a semantically related
item from the other category. In each group, the four
tasks were presented in the following order: NN, VN,
NV, and VV in half of the subjects, and in the reverse
order in the second half. The task had to be performed
within a period of 4 seconds for control subjects, and of
6 seconds for PD patients to account for the slowed
articulatory rate often present in PD. Before each task, an
example was given (e.g., for VV task, “if I tell you
‘boire’ [drink], you may answer ‘manger’ [eat] or
‘avaler’ [swallow]”).
We calculated the number of correct responses, i.e.,
semantically and grammatically adequate productions
within the given time period, accepting responses that
included a determiner followed by a noun for nouns, or
a pronoun followed by a verb for verbs. The errors were
classified as “no production” errors, grammatical errors
(i.e., production of an item semantically but not gram-
matically adequate like a verb or, more rarely, an adjec-
tive when a noun is required), semantic errors and others
(e.g., neologisms). In French, some words can be ambig-
uous in terms of grammatical category, for example
“brike” can be a conjugated or infinitive form of the verb
“briquer” [to scrub] or a name, i.e., “briquet” [lighter]. In
case of ambiguity, the response was scored as correct if
the presence of a semantic relationship helped to disen-
tangle the grammatical ambiguity. For example, in verb/
verb condition, the response “briquer” [to scrub] was
considered as correct for the target “cirer” [to polish] on
the basis of the semantic vicinity. The semantic relation-
ship was scored by two independent judges, and only
congruent judgments were considered.
Statistical Analyses
Scores were compared using analyses of variance
(ANOVAs) and planned comparisons with group, task,
and error types as factors. Nonparametric correlation
analyses were used to assess the relationships between
significantly impaired scores in PD and other variables
such as DRS and UPDRS motor scores or number of
competing lexical alternatives.
Movement Disorders, Vol. 18, No. 2, 2003
FIG. 1. Global performance (number of errors) in Parkinson’s disease
patients and control subjects on generation tasks. *P ⬍ 0.02; **P ⬍
0.008.
RESULTS
Comparisons between Groups for
Generation Performance
The scores on the four generation tasks are reported in
Figure 1. A three-factor ANOVA (group, task, and error
type) with repeated measures showed a nearly significant
main effect for group (F(1,66) ⫽ 3.58; P ⬍ 0.06), a
significant main effect for task (F(3,198) ⫽ 11.13; P ⬍
0.0014), and a significant main effect for error type
(F(3,198) ⫽ 65.01; P ⬍ 0.0001).
Planned comparisons demonstrated that controls are
significantly better than patients for VV (F(1,66) ⫽ 5.72;
P ⬍ 0.019) and NV generation tasks (F(1,66) ⫽ 7.66;
P ⬍ 0.0073). No difference was found for NN generation
(F(1,66) ⫽ 1.56; P ⬍ 0.22) and VN generation
(F(1,66) ⫽ 0.50; P ⫽ 0.48). One-factor ANOVAs per-
formed in each group showed significant differences
between tasks (control group: F(3,33) ⫽ 2.97, P ⬍ 0.04;
PD group: F(3,33) ⫽ 15.96, P ⬍ 0.001).
Post hoc pair-wise contrasts showed that, in the con-
trol group, this effect was due to the difference between
VV and NN tasks (P ⬍ 0.05). In the PD group, this effect
was due to striking differences between the two tasks,
involving the production of verbs that both elicited many
errors and the two noun-generation tasks in which per-
formances were at a normal range (VV vs. NN and NV
vs. NN, P ⬍ 0.0001; VV vs. VN, P ⬍ 0.005; NV vs. VN,
P ⬍ 0.02).
Because semantic errors and errors qualified as “oth-
ers” were very rare in the production of the subjects (less
than one error on average, all groups and tasks), we only
included “no production” and “grammatical” errors in
the analyses (see Table 2). For each type of error, we
assessed across the two groups the influence of the noun/
 DEFICIT IN VERB PROCESSING IN PD 153

TABLE 2. No production and grammatical errors on word generation tasks

Control subjects PD subjects
Noun Noun Verb Verb Noun Noun Verb Verb
Noun Verb Verb Noun Noun Verb Verb Noun
No production 3.8 5.5 6.5 4.4 2.3 6.9 9.7 3.6
⫾3.6 ⫾4.8 ⫾3.9 ⫾3.9 ⫾2.3 ⫾6.5 ⫾6.8 ⫾3.6
Grammatical error 1.3 0.3 1.4 1.6 1.7 4.6 2.6 3.7
⫾1.6 ⫾1.2 ⫾5.3 ⫾3.8 ⫾2.3 ⫾8.02 ⫾5.3 ⫾5.8

Values are expressed as mean ⫾ S.D.

PD, Parkinson’s disease.

verb category considering either the required output
(e.g., verb in verb/verb and noun/verb conditions) or the
given stimulus (e.g., noun in noun/noun and noun/verb
conditions). We performed three-factor ANOVAs (group,
required output, given stimulus) for each type of error, in
which output and stimulus were considered as repeated
factors.
For no-production errors, the three-factor ANOVA
showed a significant main effect for the given stimulus
(F(1,66) ⫽ 17.75; P ⬍ 0.0001), a significant main effect
for the required output (F(1,66) ⫽ 67.76; P ⬍ 0.0001),
and no significant main effect for the group. All subjects
(patients and controls) exhibited more errors when the
given stimulus or the required output was a verb. Nev-
ertheless, the production of verb was slightly more dif-
ficult for PD patients than for controls (F(1,66) ⫽ 3.6;
P ⫽ 0.06), whereas no significant difference was ob-
served when the given stimulus was a verb.
For grammatical errors, the three-factor ANOVA
showed only a significant main effect for the group
(F(1,66) ⫽ 5.23; P ⬍ 0.03), PD patients showing more
grammatical errors than controls. The high standard de-
viations relative to mean values led us to perform
planned comparisons after having removed outliers (ex-
cluding values beyond mean ⫹ 1 S.D.). These compar-
isons showed more grammatical errors in PD patients
than in controls in the three tasks involving verbs, the
main contrast being observed for noun/verb condition
(P ⬍ 0.0003), whereas verb/noun and verb/verb condi-
tions generated significant, although less marked, effects
(P ⬍ 0.02 and P ⫽ 0.05, respectively).
the condition that required a switch between category,
i.e., noun/verb condition and DRS scores (Spearman
rank correlation, Rho ⫽ ⫺0.451; P ⬍ 0.0096, see Fig.
2), whereas no significant correlation was found for
verb/verb condition (P ⫽ 0.34). Because the noun/verb
condition showed correlation between DRS and error
rate, we further explored this correlation by testing the
relationship between error rate and subscores of the
DRS. We found a significant correlation between the
error rate and the memory subscore (P ⬍ 0.005) and a
trend between the error rate and the conceptualisation
subscore (P ⫽ 0.088).
Second, we sought to relate motor impairment mea-
sured by UPDRS motor scores and impaired perfor-
mance on N/V and V/V conditions insofar as only action
verbs were included in the protocol. We failed to find
significant correlations between UPDRS motor scores
and error rate for verb/verb (Spearman rank correlation,
Rho ⫽ 0.156; P ⬍ 0.44), whereas a tendency was ob-
served for noun/verb (Spearman rank correlation, Rho ⫽
351; P ⬍ 0.09).
Correlation between Competing Alternatives and
Performance in PD Group
Because our generation tasks required subjects to gen-
erate a verb or a noun in response to a noun or a verb, the

Correlation between Cognitive and Motor Status

and Performance in PD Group
We explored first in our PD patients whether a possi-
ble slight general cognitive impairment, reflected by the
range of the DRS scores might have influenced the two
impaired generation performance, i.e., verb/verb and
noun/verb generation. We then performed nonparametric
correlations between DRS and error rate. We found a FIG. 2. Correlation between DRS score and NV errors. DRS, Demen-
significant negative correlation between performance on tia Rating Scale; NV, noun-to-verb generation.

Movement Disorders, Vol. 18, No. 2, 2003

154 P. PÉRAN ET AL.
given stimuli can provide high or low selection con-
straint for the required output choice and the degree of
competition for selection between different responses
can account for abnormal performance (see Thompson-
Schill et al.26). For example, in verb/verb generation task,
the item “allumer” [to switch on a light] exemplifies high
selection constraint because 68% of the controls pro-
duced the same response, “éteindre,” [to switch off a
light], the second most common response being
“éclairer” [to lighten] produced by 21% of the controls.
Conversely, the item “coudre” [to sew] is typical of low
selection constraint with a variety of responses like “dé-
coudre” [to rip a seam], “tricoter” [to knit], or “repriser”
[to mend], each one eliciting roughly 10% of responses.
We then calculated for each task, on the basis of
controls’ correct responses, a ratio of the relative fre-
quency of the most common response to the relative
frequency of the second-most common response as a
measure of response strength (ratio based on the method
described by Thompson-Schill et al.26). For example, the
ratio for “allumer” is 68/21 ⫽ 3.23, and the ratio for
“coudre” is 15/12 ⫽ 1.25.
We assessed the influence of increased selection de-
mand on performance in PD group for the two tasks that
induced significantly poor performance, i.e., V/V and
N/V generation. We failed to find any significant corre-
lation between the computed ratio and the mean global
error rate in the two tasks (Spearman rank correlation,
Rho ⫽ ⫺0.179; P ⬍ 0.27 for V/V, Rho ⫽ 0.142; P ⬍
0.38 for N/V).
DISCUSSION
This study evidences a clear language deficit in non-
demented PD patients in tasks designed to explore verb
processing. Considering the global scores on the four
generation tasks, nondemented PD patients were signif-
icantly more impaired in verb-generation tasks (V/V and
N/V) than controls, whereas their scores were similar to
those of controls in noun-generation tasks (V/N and
N/N). The analysis of error type gave rise to a twofold
pattern of results. First, with regard to the absence of
response, the conditions including verb processing led to
difficulties for the two groups, but these difficulties were
more prominent in PD patients than in controls in the
tasks requiring verb production. Second, considering
grammatical errors, PD patients produced more errors
than controls but only in tasks involving verbs, with a
pronounced deficit in the noun/verb condition, a task that
requires one to shift between lexical categories (cross-
category task).
The first interpretation for these findings is that word-
generation tasks might be globally more difficult for PD
Movement Disorders, Vol. 18, No. 2, 2003
patients, especially because of the stress imposed by the
time constraints. The absence of significant difference
between performance of PD patients and controls on
noun-generation tasks, however, seems to exclude any
influence of “global difficulty” in processing word gen-
eration. Nevertheless, to avoid the influence of the “un-
specific slowness” induced by PD, patients were allo-
cated a longer time to produce responses than controls.
Although this temporal advantage could account for the
normal performance on noun generation, it reinforces the
significance of poor performance observed on verb
generation.
This deficit does not seem to reflect a mere accentua-
tion of a difficulty that would also be observed in con-
trols for verb processing. Indeed, it has been suggested
that verbs are inherently more difficult to process than
nouns due to their greater syntactic and/or semantic
complexity,27 and that verb/verb generation task appears
not to be particularly easy even for controls, as shown by
their significant poorest performance on this task. How-
ever, this slight difficulty of control subjects is not found
in the noun/verb generation task (in which performance
is similar to that observed in noun-generation tasks); in
terms of performance, this is contrary to the results of the
PD patients, in whom the two tasks requiring verb pro-
duction differ drastically from the two tasks requiring
noun production.
The difference between noun and verb generation can
be interpreted as an actual language deficit for verbs in
nondemented PD patients who are considered to present
a mild impairment of frontal functions. This deficit is not
related to a lexical factor such as a high selection demand
for some verbs in the tasks, as shown by the lack of
correlations between competing alternatives and perfor-
mance in our group of patients.
This verb deficit is congruent with that described by
Grossman and colleagues,28 who showed a verb-learning
impairment in a group of nondemented PD patients.
Moreover, these results could evoke the deficit of verb
naming observed in patients with frontotemporal demen-
tia.29 Our findings are also congruent with those showing
a selective impairment of verb processing in motor neu-
rone disease associated with pathological changes in two
frontal areas, namely Brodmann areas 44 and 45.30 This
latter result confirms the frontal localization observed in
neuroimaging studies for verb representations with the
implication of Brodmann areas 44, 45, and 47.13–19
Two lines of interpretation can be proposed to account
for the verb deficit observed in our group of PD patients.
First, a relationship between verb processing and premo-
tor cortex has been conceptualised as related to repre-
sentations of action.9 The deficit of verb production in
 DEFICIT IN VERB PROCESSING IN PD
PD patients could correspond to an impairment of ac-
cessing the linguistic and mental representations of ac-
tion, an impairment that reflects the frontal dysfunction
in these patients. Indeed, functional neuroimaging stud-
ies have suggested that not only actual motor perfor-
mance but also mental representations of actions recruit
neural activities in the primary and association motor
areas (for review, see Decety31). Moreover, a functional
deficit of the medial premotor cortex has been shown in
motor activation imaging studies in PD patients.32,33 In
our study, the lack of significant correlations between
verb production conditions and motor performance does
not seem to comfort the motor hypothesis. However, the
UPDRS offers only a gross evaluation of motor distur-
bances. This assessment of motor performance concerns
only meaningless gestures. Moreover, it focuses mainly
on limb rather than axial movements that are likely to be
more related to impairment of speech production. Limb
and axial motor disorders are only weakly related and
may have specific neural substrates. Indeed, a recent
study34 has shown in normal subjects different brain
activity sources for verbs referring to actions executed
with the legs or the face.
The second line of interpretation relates to the hypoth-
esis of a grammatical deficit in PD. Indeed, neuropsy-
chological studies of PD patients have shown significant
impairments in sentence comprehension,35,36 and Gross-
man and associates28 have shown that nondemented PD
patients experienced difficulties in appreciating gram-
matical information when they had to learn a new verb.
Moreover, syntactic and morphosyntactic processing has
been addressed by several functional neuroimaging stud-
ies that showed consistent activations in the left inferior
frontal cortex in normal subjects.37–39 A recent study by
Shapiro and coworkers40 using repetitive transcranial
magnetic stimulation suggested that the left prefrontal
cortex is selectively engaged in producing verbs as gram-
matical objects. It is then possible to assume that the
deficit for verb production we observed is related to a
grammatical impairment as a consequence of a dysfunc-
tion of the frontal cortex in nondemented PD patients.
This interpretation seems corroborated by the significant
presence of grammatical errors in the tasks requiring
verb processing in our group of patients.
Moreover, in addition to a grammatical difficulty
linked to verb production, one may assume that subtle
cognitive deficits might hamper lexical retrieval and pro-
duction in some patients, even though there was no
evidence of dementia in any of them. Indeed, we ob-
served a relationship between the DRS score and the
performance on the N/V generation task, that requires
shifting from a grammatical category to another. Shifting
155
processes demands cognitive flexibility, a process that is
under frontal lobe control and, consequently, tends to be
reduced in patients with frontostriatal dysfunction, in-
cluding demented PD patients41 and patients with frontal
focal lesions.42 Although performance was not statisti-
cally different from that of controls in the other “shift-
ing” task, namely the V/N generation, a correlation be-
tween the DRS score and the performance was also
observed in this task (Rho ⫽ ⫺0.525; P ⬍ 0.003). The
analysis of correlation between error rate and DRS sub-
scores did not show a clear-cut profile, suggesting a
massive link between impairment in shifting generation
tasks and executive dysfunction. Rather, we observed a
correlation between generation task performance and the
DRS memory subscore. In the absence of extensive
testing of memory and executive functions, it is difficult
to further speculate on this finding. Nevertheless, when
comparing global scores or rate of grammatical errors
between PD and controls, we found a significant differ-
ence for noun/verb generation but not for the verb/noun
generation. This discrepancy may be accounted for by
the convergence of two sources of deleterious effects in
the noun/verb task, i.e., shifting process and production
of verb, whereas in the verb/noun task, the impact of
shifting is likely to be compensated by the easier noun
production context.
In conclusion, our findings suggest that nondemented
PD patients may present a particular difficulty to produce
verbs in a word generation situation. Further studies,
including neuroimaging activation experiments, might
contribute to our understanding of the frontal dysfunc-
tion induced by early Parkinson’s disease, and also of the
two alternative interpretations of this deficit, namely the
“motor” theory (verbs as action representations) versus
the “grammatical” theory (verbs as lexical category).
Acknowledgments: This study was supported by grants
from MENRT “Cognitique-Action 2000” and is part of the
Langage, Image, Movement and Cognition project.
REFERENCES
1. Dubois B, Pillon B. Cognitive deficits in Parkinson’s disease.
J Neurol 1997;244:2– 8.
2. Hanley J, Dewick H, Davies A, Playfer J, Turnbull C. Verbal
fluency in Parkinson’s disease. Neuropsychologia 1990;28:737–
741.
3. Bayles K, Trosset M, Tomoeda C, Montgomery EJ, Wilson J. Gen-
erative naming in Parkinson disease patients. J Clin Exp Neuro-
psychol 1993;15:547–562.
4. Troster AI, Fields JA, Testa JA, Paul RH, Blanco CR, Hames KA,
Salmon DP, Beatty WW. Cortical and subcortical influences on
clustering and switching in the performance of verbal fluency
tasks. Neuropsychologia 1998;36:295–304.
5. Piatt A, Fields J, Paolo A, Koller W, Troster A. Lexical, semantic,
and action verbal fluency in Parkinson’s disease with and without
dementia. J Clin Exp Neuropsychol 1999;21:435– 443.
Movement Disorders, Vol. 18, No. 2, 2003
156 P. PÉRAN ET AL.
6. Auriacombe S, Grossman M, Carvell S, Stern M, Hurtig HI.
Verbal fluency deficits in Parkinson’s disease. Neuropsychology
1993;7:182–192.
7. Zingeser LB, Berndt RS. Retrieval of nouns and verbs in agram-
matism and anomia. Brain Lang 1990;39:14 –32.
8. Caramazza A, Hillis A. Lexical organization of nouns and verbs in
the brain. Nature 1991;349:788 –790.
9. Damasio A, Tranel D. Nouns and verbs are retrieved with differ-
ently distributed neural systems. Proc Natl Acad Sci U S A
1993;90:4957– 4960.
10. Breedin SD, Saffran EM, Schwartz MF. Semantic factors in verb
retrieval: an effect of complexity. Brain Lang 1998;63:1–31.
11. Daniele A, Giustolisi L, Silveri MC, Colosimo C, Gainotti G.
Evidence for a possible neuroanatomical basis for lexical process-
ing of nouns and verbs. Neuropsychologia 1994;32:1325–1341.
12. Tyler LK, Russell R, Fadili J, Moss HE. The neural representation
of nouns and verbs: PET studies. Brain 2001;124:1619 –1634.
13. Petersen SE, Fox PT, Posner MI, Mintun M, Raichle ME. Positron
emission tomographic studies of the processing of single words. J
Cogn Neurosci 1989;1:153–170.
14. Wise R, Chollet F, Hadar U, Friston K, Hoffner E, Frackowiak R.
Distribution of cortical neural networks involved in word compre-
hension and word retrieval. Brain 1991;114:1803–1817.
15. Raichle M, Fiez J, Videen T, MacLeod A, Pardo J, Fox P, Petersen
S. Practice-related changes in human brain functional anatomy
during nonmotor learning. Cereb Cortex 1994;4:8 –26.
16. Martin A, Haxby JV, Lalonde FM, Wiggs CL, Ungerleider LG.
Discrete cortical regions associated with knowledge of color and
knowledge of action. Science 1995;270:1102–1105.
17. Warburton E, Wise R, Price C, Weiller C, Hadar U, Ramsay S,
Frackowiak RS. Noun and verb retrieval by normal subjects.
Studies with PET. Brain 1996;119:159 –179.
18. Tatsumi IF, Fushimi T, Sadato N, Kawashima R, Yokoyama E,
Kanno I, Senda M. Verb generation in Japanese. A multicenter
PET activation study. Neuroimage 1999;9:154 –164.
19. Perani D, Cappa S, Schnur T, Tettamanti M, Collina S, Rosa M,
Fazio F. The neural correlates of verb and noun processing. A PET
study. Brain 1999;122:2337–2344.
20. Gibbs WR, Lees AJ. The relevance of the Lewy body to the
pathogenesis of idiopathic Parkinson’s disease. J Neurol Neuro-
surg Psychiatry 1988;51:745–752.
21. Fahn S, Elton RL, and the members of the UPDRS Development
Committee. Unified Parkinson’s disease rating scale. In: Fahn S,
Marsden CD, Goldstein M, Calne DB, editors. Recent develop-
ments in Parkinson’s disease. Vol. II. Florham Park, NJ: Macmil-
lan; 1987. p 153–163.
22. Folstein M, Folstein S, Mchugh P. Mini-mental state. A practical
method for grading the cognitive state of patients for the clinician.
J Psychiatr Res 1975;12:189 –198.
23. Mattis S. Dementia rating scale. Odessa: Psychological Assess-
ment Resources: 1988.
24. Baudot J. Fréquences d’utilisation des mots en français contempo-
rain. Montréal: Presses de l’Université de Montréal; 1992.
Movement Disorders, Vol. 18, No. 2, 2003
25. Cardebat D, Candelon N, Kuznierek L, Viallard G, Démonet JF,
Lambert J. La génération de mots isolés chez les sujets normaux.
Acta Neurol Belg 2000;100:24 –33.
26. Thompson-Schill SL, Swick D, Farah MJ, D’Esposito M, Kan IP,
Knight RT. Verb generation in patients with focal lesions: a
neuropsychological test of neuroimaging findings. Proc Natl Acad
Sci U S A 1998;95:15855–15860.
27. Gentner D. Some interesting differences between verbs and nouns.
Cogn Brain Theory 1981;4:161–178.
28. Grossman M, Stern MB, Gollomp S, Vernon G, Hurtig HI. Verb
learning in Parkinson’s disease. Neuropsychology 1994;8:413–
423.
29. Cappa SF, Binetti G, Pezzini A, Padovani A, Rozzini L, Trabucchi
M. Object and action naming in Alzheimer’s disease and fronto-
temporal dementia. Neurology 1998;50:351–355.
30. Bak TH, O’Donovan DG, Xuereb JH, Boniface S, Hodges JR.
Selective impairment of verb processing associated with patholog-
ical changes in Brodmann areas 44 and 45 in the motor neurone
disease-dementia-aphasia syndrome. Brain 2001;124:103–120.
31. Decety J. The neurophysiological basis of motor imagery. Behav
Brain Res 1996;77:45–52.
32. Playford ED, Jenkins IH, Passingham RE, Nutt J, Frackowiak RS,
Brooks DJ. Impaired mesial frontal and putamen activation in
Parkinson’s disease: a positron emission tomography study. Ann
Neurol 1992;32:151–161.
33. Sabatini U, Boulanouar K, Fabre N, Martin F, Carel C, Colonnese
C, Bozzao L, Berry I, Montastruc JL, Chollet F, Rascol O. Cortical
motor reorganization in akinetic patients with Parkinson’s disease:
a functional MRI study. Brain 2000;123:394 – 403.
34. Pulvermuller F, Harle M, Hummel F. Neurophysiological distinc-
tion of verb categories. Neuroreport 2000;11:2789 –2793.
35. Lieberman P, Kako E, Friedman J, Tajchman G, Feldman LS,
Jiminez EB. Speech production, syntax comprehension, and cog-
nitive deficits in Parkinson’s disease. Brain Lang 1992;43:169 –
189.
36. Grossman M. Sentence processing in Parkinson’s disease. Brain
Cogn 1999;40:387– 413.
37. Caplan D, Alpert N, Waters G. Effects of syntactic structure and
propositional number on patterns of regional cerebral blood flow.
J Cogn Neurosci 1998;10:541–552.
38. Caplan D, Alpert N, Waters G. PET studies of syntactic processing
with auditory sentence presentation. Neuroimage 1999;9:343–351.
39. Moro A, Tettamanti M, Perani D, Donati C, Cappa SF, Fazio F.
Syntax and the brain: disentangling grammar by selective anoma-
lies. Neuroimage 2001;13:110 –118.
40. Shapiro K, Pascual-Leone A, Mottaghy F, Gangitano M, Car-
amazza A. Grammatical distinctions in the left frontal cortex. J
Cogn Neurosci 2001;13:713–720.
41. Troyer AK, Moscovitch M, Winocur G, Leach L, Freedmann M.
Clustering and switching on verbal fluency tests in Alzheimer’s
disease and Parkinson’s disease. J Int Neuropsychol Soc 1998;4:
137–143.
42. Troyer AK, Moscovitch M, Winocur G, Alexander M, Stuss D.
Clustering and switching on verbal fluency: the effects of focal frontal
and temporal-lobe lesions. Neuropsychologia 1998;36:499–504.