Elsevier

NeuroImage

Volume 25, Issue 2, 1 April 2005, Pages 588-599
NeuroImage

Executive dysfunction in Parkinson's disease is associated with altered pallidal–frontal processing

https://doi.org/10.1016/j.neuroimage.2004.11.023Get rights and content

Abstract

Executive dysfunction in Parkinson's disease is well documented, but it is still unclear whether this results from (i) prefrontal dysfunction, (ii) striatal dysfunction, or (iii) altered striatal outflow to the prefrontal cortex. To clarify this issue, we used H215O PET to asses six nondemented and nondepressed patients with Parkinson's disease and six matched controls while they performed a task involving executive function, random number generation (RNG), and a control counting task. To assess the effect of increasing task demands, each task was performed at three rates. Both groups showed significant increase in nonrandomness of responses during RNG at faster rates, which was differentially greater for the patients at the faster rate. The controls showed significant activation of the lateral and medial prefrontal cortex and superior and medial parietal cortex during RNG relative to counting. For the same comparison, the patients did not show any activity in medial frontal structures. The controls showed significantly greater mesial frontotemporal activation during counting than RNG, whereas the patients did not show any modulation of regional cerebral blood flow (rCBF) in these areas with task. With faster rates of RNG, the controls showed rCBF increase in the right internal segment of globus pallidus (GPi) and a decrease in frontal cortex. The patients showed the opposite pattern of subcortical and frontal rCBF change with faster rates. The results suggest that executive dysfunction in Parkinson's disease is associated with a failure to modulate frontal activation with increased task demands (nature of task or rate), a deficit associated with altered rCBF in the GPi, the final basal ganglia output pathway to frontal cortex rather than any intrinsic prefrontal dysfunction.

Introduction

It is now well documented that the cognitive profile of patients with Parkinson's disease is characterized by executive dysfunction (e.g., Gotham et al., 1988, Owen et al., 1997, Taylor et al., 1986). Executive function is considered to be under the control of the prefrontal cortex (Norman and Shallice, 1986), which is reciprocally connected with the striatum (Alexander and Crutcher, 1990). From neuropsychological evidence, it is unclear whether the executive deficits documented in Parkinson's disease are a result of (i) abnormal prefrontal processing, (ii) abnormal striatal functioning, or (iii) altered basal ganglia output to the prefrontal cortex. While a number of imaging studies have used cognitive activation paradigms (Cools et al., 2002, Dagher et al., 2001, Lewis et al., 2003, Mattay et al., 2002, Owen et al., 1998) to investigate the mechanisms of executive dysfunction in Parkinson's disease, they have also failed to clarify this issue. In fact, in a recent review of functional imaging of cognition in Parkinson's disease, Carbon and Marié (2003) noted that “the specific contributions of mesocortical dopamine depletion and striatal dysfunction…remain to be separated”. An example of this lack of clarity in the results of imaging studies is that while some have found that when performing the Tower of London or spatial working memory tasks that engage executive processes, patients with Parkinson's disease show significant differences from controls in subcortical (globus pallidus (GPi) or caudate) activation (Dagher et al., 2001, Owen et al., 1998), on the same tasks others have found that patients differed from controls only in frontal activation (Cools et al., 2002). Therefore, while the proposal that executive dysfunction in Parkinson's disease may result from altered basal ganglia output to the prefrontal cortex is not new (e.g., Taylor et al., 1986), no clear empirical support for this is available at the moment. The first aim of our study was to compare patterns of brain activation in patients with Parkinson's disease compared to elderly controls during the performance of random number generation (RNG), a cognitive task that involves a number of executive processes, such as internal response generation through suppression of habitual counting, monitoring, and switching production strategies.

We have conducted a number of clinical, imaging, and transcranial magnetic stimulation studies of RNG (Brown et al., 1998, Jahanshahi et al., 1998, Jahanshahi et al., 2000b). In these studies, we have found indices of habitual counting, particularly ‘counting in ones’, to be a measure of the nonrandomness of the subject's output that is sensitive to capacity manipulations. During RNG, patients with Parkinson's disease show a bias toward habitual counting. When RNG is performed under more attention-demanding dual task conditions concurrently with a manual tracking task, controls show a reversal of their response bias that becomes marked by a significant increase in ‘counting in ones’ and the patients with Parkinson's disease show a further magnification of ‘counting in ones’ (Brown et al., 1998). We have proposed a network modulation model of RNG, which suggests that in order to generate numbers in a random fashion, subjects have to suppress habitual counting. This suppression of habitual counting is achieved through the modulatory (inhibitory) control of the left DLPFC over a number associative network in the temporal cortex (Jahanshahi et al., 1998). Support for the model has come from a PET activation study of RNG in young normals (Jahanshahi et al., 2000b). RNG was associated with increased regional cerebral blood flow (rCBF) in the left DLPFC and decreased blood flow in the temporal cortex relative to a counting task. Also, as predicted by the model, rCBF in the left DLPFC showed a negative covariation with blood flow in the temporal cortex and with behavioral measures of habitual counting. We have also shown that repetitive TMS over the left DLPFC increases habitual counting during RNG (Jahanshahi et al., 1998) and stereotyped responding (e.g., ABC or BBC) during a random letter generation task (Jahanshahi and Dirnberger, 1999), indicating that this area plays an essential role in the suppression of habitual or stereotyped responses during random generation.

Previous studies have established that deficits shown by patients with Parkinson's disease are magnified when task demands increase (e.g., Brown and Marsden, 1988, Brown and Marsden, 1991, Brown et al., 1993, Owen et al., 1993, Owen et al., 1997). With random generation tasks, when the processing demands are increased, for example, by performing the task at faster rates, habitual or stereotyped responses also increase and the output becomes more nonrandom (Baddeley, 1966, Jahanshahi et al., 2000b, Robertson et al., 1996). In our previous PET study of RNG in young normals, faster rates of RNG were associated with greater nonrandomness of responses, decreased rCBF in the left and right DLPFC and superior parietal cortex, and increased rCBF in the left superior temporal cortex (Jahanshahi et al., 2000b). The rate dependency of brain activation during RNG has been confirmed with fMRI (Daniels et al., 2003). Performance of RNG under dual task conditions such as concurrently with a sorting task also increases task demands and results in a greater nonrandomness of the output in normals (Baddeley, 1966, Brown et al., 1998, Robertson et al., 1996) as well as in patients with Parkinson's disease (Brown et al., 1998). In light of the results of these behavioral studies, our second aim was to examine the effects of performing RNG at faster rates that increase processing demands on patterns of brain activation in patients with Parkinson's disease relative to healthy participants.

Based on previous results, we made a number of specific predictions.

  • 1.

    Relative to healthy controls, patients with Parkinson's disease would show greater increase in habitual counting at faster rates of RNG.

  • 2.

    Relative to healthy controls, patients with Parkinson's disease would show impaired activation of frontal particularly medial frontal structures and the basal ganglia during RNG relative to a control counting task.

  • 3.

    Relative to healthy controls, patients with Parkinson's disease would fail to modulate frontal activation with greater demands of performing RNG at faster rates.

Section snippets

Participants

Six patients (two male, four female) with a mean age of 60.1 years (SD 12.2) with a clinical diagnosis of Parkinson's disease according to the criteria of the UK Parkinson's disease Brain Bank (Hughes et al., 1992) and with presence of at least two of the symptoms of tremor, akinesia, and rigidity participated in the study. All were being treated with dopaminergic medication that was effective in controlling the symptoms. The mean Hoehn and Yahr (1967) rating was 3.0 (SD = 1.3). The mean

Results

The two groups were matched in terms of age, scores on the Mini Mental, and estimates of verbal IQ obtained from the NART (P > 0.05). The patients had significantly higher scores on the BDI than the controls (t = 4.37, df = 10, P < 0.01). However, none of the patients had scores above 14, indicative of mild, moderate, or severe depression (Beck et al., 1961).

Discussion

The patients and controls were matched in terms of age and NART estimates of premorbid IQ. The patients did not show any evidence of cognitive deficit. Although self-reported BDI scores were higher for patients than elderly controls, none of the patients were depressed. The patients and controls did not differ in terms or the number of responses produced at the three rates for either the RNG or COUNT tasks, showing that both groups were able to synchronize their responses with the pacing tone

Acknowledgments

The financial support of the Wellcome Trust (GD, MJ, CDF) is gratefully acknowledged.

References (52)

  • C. Robertson et al.

    The effects of Parkinson's disease on the capacity to generate information randomly

    Neuropsychologica

    (1996)
  • G.E. Alexander et al.

    Parallel organization of functionally segregated circuits linking basal ganglia and cortex

    Annu. Rev. Neurosci.

    (1986)
  • A.D. Baddeley

    The capacity for generating information by randomisation

    Q. J. Exp. Psychol.

    (1966)
  • A.T. Beck et al.

    An inventory for measuring depression

    Arch. Gen. Psych.

    (1961)
  • K. Brodmann

    Vergleichende Lokalisationslehre der Grosshirnrinde in ihren Prinzipien dargestellt aufgrund des Zellenbaues

    (1909)
  • R.G. Brown et al.

    Internal versus external cues and the control of attention in Parkinson's disease

    Brain

    (1988)
  • R.G. Brown et al.

    Dual task performance and processing resources in normal subjects and patients with Parkinson's disease

    Brain

    (1991)
  • R.G. Brown et al.

    Response choice in Parkinson's disease

    Brain

    (1993)
  • M. Carbon et al.

    Functional imaging of cognition in Parkinson's disease

    Curr. Opin. Neurol.

    (2003)
  • M.-J. Catalan et al.

    A PET study of sequential finger movements of varying length in patients with Parkinson's disease

    Brain

    (1999)
  • R.R. Cools et al.

    Dopaminergic modulation of high-level cognition in Parkinson's disease: the role of the prefrontal cortex revealed by PET

    Brain

    (2002)
  • A. Dagher et al.

    The role of the striatum and hippocampus in planning. A PET activation study in Parkinson's disease

    Brain

    (2001)
  • K.J. Friston et al.

    The relationship between global and local changes in PET scans

    J. Cereb. Blood Flow Metab.

    (1990)
  • K.J. Friston et al.

    Statistical parametric maps in functional imaging: a general approach

    Hum. Brain Mapp.

    (1995)
  • C.D. Frith

    The role of the dorsolateral prefrontal cortex in the selection of action, as revealed by functional imaging

  • A.M. Gotham et al.

    “Frontal” cognitive function in patients with Parkinson's disease “on” and “off” levodopa

    Brain

    (1988)
  • Cited by (68)

    • Alterations in white matter fiber in Parkinson disease across different cognitive stages

      2022, Neuroscience Letters
      Citation Excerpt :

      For the ANOVA analysis with significant areas across multiple groups, we carried out the Pearson correlation analysis between the fixel values and clinical scales, which is further corrected by FDR. In combining clinical scales, we did not limit to the aspect of cognitive function but also focused on motor and emotional processes, since motor functions (i.e., learning, planning, and execution), and emotional changes such as depression are cognition dependent [25,26]. Participants underwent a series of tests evaluating various functions, including (1) UPDRS-III as a measure of motor function; (2) MOCA for cognitive function; (3) Geriatric Depression Scale-Short Form (GDS) for possible symptoms of depression; (4) Hopkins Verbal Learning Test-Revised (HVLT-R) as a measure of verbal memory, including HVLT-R discrimination (HVLT-R-DIS), HVLT-R immediate recall (HVLT-R-IR), and HVLT-R retention (HVLT-R-RET); (5) Semantic Fluency test (SFT) for reflecting executive function; (6) WMS-III Letter Number Sequencing (LNS) test as a measure of working memory; (7) State-Trait Anxiety Inventory for Adults (STAI) in order to show the psychological condition; (8) Symbol Digit Modalities Test (SDMT) for assessing attention function; (9) REM Sleep Behavior Disorder Questionnaire (RBDSQ) assessment as a measure of sleep behavior (https://www.ppmi-info.org/wp-content/uploads/2010/04/PPMI-General-Operations-Manual.pdf).

    • Functional neuroanatomy of cognition in Parkinson's disease

      2022, Progress in Brain Research
      Citation Excerpt :

      Mortimer et al. (1982) found a significant correlation between the severity of the motor symptoms and the degree of cognitive deficits of Parkinson's patients. This functional correlation may suggest that the same mechanisms underlying the motor symptoms of PD can cause or contribute to cognitive dysfunctions, an idea that has received more supporting evidence in later studies (Dirnberger et al., 2005; Dubois and Pillon, 1997; Gotham et al., 1988). In sum, substantial evidence exists to suggest that executive dysfunction in PD is due to dopaminergic depletion in the striatum disrupting transmission in the frontostriatal network (Dirnberger et al., 2005; Dubois and Pillon, 1997; Gotham et al., 1988).

    View all citing articles on Scopus
    1

    Current address: Department of Neurology, University of Vienna, Austria.

    View full text