Research ReportPrefrontal cortex and striatal activation by feedback in Parkinson's disease
Introduction
In everyday life positive feedbacks in response to human action are an obvious phenomenon. These feedbacks may differ both in style and intensity, e.g. receiving money will be evaluated differently than a positive verbal remark. Positive feedbacks (also designated as “rewards”) reinforce goal-directed behavior and evoke positive feelings (Schultz, 1997). Various brain areas may be associated with positive feedback processing (reward processing) in humans, in dependence of the behavioral task used during neuroimaging. However, certain brain areas are consistently associated with reward processing in healthy humans: striatum (nucleus accumbens, nucleus caudatus, putamen), orbitofrontal cortex, anterior cingulate gyrus and other parts of the prefrontal cortex (McClure et al., 2004).
Parkinson's disease (PD) is characterized by a dysfunctional striatum, due to a progressive degeneration of dopaminergic neurons in the substantia nigra. Motor symptoms are the clinical hallmark, however depression, cognitive dysfunction and other signs and symptoms are often present (Lauterbach, 2004). The striatum is extensively connected with the prefrontal cortex through the so-called fronto-striatal circuits. Due to the dysfunction of the striatum these circuits are functionally impaired as well in PD (Alexander et al., 1986).
Neuroimaging studies investigated reward processing in PD using various tasks and methods. Using a pattern recognition task during H215O Positron Emission Tomography (PET), Kunig et al. (2000) showed a lack of striatal activation in response to reward in PD patients. Furthermore, they reported that the dorsolateral prefrontal cortex and the anterior cingulate gyrus, did show activation in PD patients in response to reward, suggesting a compensatory strategy reflected by a shift from striatal to prefrontal regions for reward processing. Schott et al. (2007) also reported such increased cortical activation. They used a reward-prediction paradigm during functional Magnetic Resonance Imaging (fMRI) and showed increased activation during reward feedback in the medial prefrontal cortex and anterior cingulate gyrus in PD patients. Goerendt et al. (2004) examined the processing of rewards of different magnitudes in PD patients and healthy controls using H215O PET. They showed that task completion time improved with increasing reward magnitude. In healthy controls this was associated with increased prefrontal and rhinal cortex activity. In contrast, PD patients showed a cerebellar vermis overactivation, also suggesting a compensatory strategy which in that study was reflected by a shift from prefrontal regions to the cerebellum. These studies thus provide arguments that reward processing is altered in PD.
Rewards or feedback also play an important role in learning. It has been reported that PD patients specifically show impairments when learning is based upon feedback compared to learning without receiving feedback (Shohamy et al., 2004). Furthermore, PD patients “on” levodopa learn better when learning is based upon receiving positive feedback, while PD patients “off” levodopa learn better when learning is based upon receiving negative feedback (Frank et al., 2004). Learning is thus influenced by the use of levodopa.
The present study uses fMRI to investigate the consequences of dysfunctional fronto-striatal circuits for the processing of positive feedback, using PD as a model disease. In this context not many studies have been performed. The current study aims to extend previous findings by investigating two types of positive feedback (“monetary feedback” and “positive informative feedback”) rather than rewards of different magnitudes. Both types are compared to a “meaningless feedback” condition. All three types are embedded in a working memory task.
Since PD is characterized by a dysfunction of the fronto-striatal circuits, we specifically focused on the brain activation patterns in the basal ganglia and frontal regions.
In addition, we have addressed the question whether positive informative feedback is just as salient as monetary feedback by comparing the activation in the striatum and prefrontal cortex during the positive informative feedback condition to the activation within these areas during the monetary feedback and meaningless feedback conditions. This was investigated in PD patients “on” levodopa compared to elderly healthy controls. We hypothesized that PD patients show an altered feedback processing. Specifically, it was expected that PD patients show a decreased striatal activation during both types of positive feedback conditions compared to the meaningless feedback condition, while showing increased prefrontal activation during both types of positive feedback conditions compared to the meaningless feedback condition.
Section snippets
Voxel-based analysis
In this section differences are shown that resulted from both within and between group analyses, contrasting the monetary feedback condition to the meaningless feedback condition, which was used as input for the ROI analysis. This enabled the identification of the prefrontal and striatal foci of maximal intensity.
Within the elderly healthy controls, the monetary feedback condition versus meaningless feedback condition contrast, yielded significant bilateral high parietal, a large pons/midbrain
Discussion
The aim of this study was to investigate the consequences of dysfunctional fronto-striatal circuits for the processing of two different types of positive feedback conditions (monetary feedback and positive informative feedback) both compared to a meaningless feedback condition. This was investigated in PD patients “on” levodopa compared to elderly healthy controls. In addition, positive informative feedback was compared to monetary feedback to determine whether positive informative feedback was
Subjects
Twenty-three subjects were included in this study: eleven PD patients (nine males (82%) and two females (18%)) and twelve healthy controls (ten males (83%) and two females (17%)). PD patients were on average 58.7 (SD = 7.6) years old and showed a mean score of 5.9 (SD = 0.8) on a Dutch education scale ranging from 1 (Elementary school not finished) to 7 (University degree). Demographic and illness characteristics of all patients are described in Table 3. Healthy controls were on average 58.8 (SD =
Acknowledgments
Partial financial support for this study was obtained from the Prinses Beatrix Foundation (PBF). Dr. Remco Renken holds a Hazewinkel-Behringer fellowship. We are also especially grateful to Dr. W. Schultz for allowing us to use his task paradigm.
References (30)
- et al.
Gender differences in limbic responsiveness, by SPECT, following a pharmacologic challenge in healthy subjects
Neuroimage
(2003) - et al.
Cortical 6-[18F]fluoro-l-dopa uptake and frontal cognitive functions in early Parkinson's disease
Neurobiol. Aging
(2005) - et al.
Nigrostriatal dopamine system and motor lateralization
Behav. Brain Res.
(2000) - et al.
Placebo mechanisms and reward circuitry: clues from Parkinson's disease
Biol. Psychiatry
(2004) - et al.
Gender differences in the mesocorticolimbic system during computer game-play
J. Psychiatr. Res.
(2008) - et al.
FMRI visualization of brain activity during a monetary incentive delay task
Neuroimage
(2000) The neuropsychiatry of Parkinson's disease and related disorders
Psychiatr. Clin. North Am.
(2004)- et al.
Reward mechanisms in the brain and their role in dependence: evidence from neurophysiological and neuroimaging studies
Brain Res. Brain Res. Rev.
(2001) Dopamine neurons and their role in reward mechanisms
Curr. Opin. Neurobiol.
(1997)- et al.
Cerebral activation related to implicit sequence learning in a Double Serial Reaction Time task
Brain Res.
(2006)
Parallel organization of functionally segregated circuits linking basal ganglia and cortex
Annu. Rev. Neurosci.
Region of interest analysis using an SPM toolbox
Neuroimage
Expectation and dopamine release: mechanism of the placebo effect in Parkinson's disease
Science
Unilateral pallidotomy versus bilateral subthalamic nucleus stimulation in PD: a randomized trial
Neurology
Unified Parkinsons disease rating scale
Cited by (12)
Subjective and objective assessment of executive functions in Parkinson's disease
2011, Journal of the Neurological SciencesCitation Excerpt :This domain involves emotional and reward processing necessary for appropriate adaptive responding in the absence of cognitive analysis, habit or environmental cues [14]. Problems with emotional and reward processing have also been described in PD [15–18]. Since PD patients reported the same number of problems as their relatives it can be assumed that they had a good insight into their daily life functioning.
Why gamblers fail to win: A review of cognitive and neuroimaging findings in pathological gambling
2010, Neuroscience and Biobehavioral ReviewsImpulse control disorders in Parkinson’s disease
2018, Journal of Neural TransmissionCortical and striatal reward processing in Parkinson's disease psychosis
2017, Frontiers in NeurologyPutamen activation represents an intrinsic positive prediction error signal for visual search in repeated configurations
2016, Open Neuroimaging JournalPsychosocial modulators of motor learning in parkinson’s disease
2016, Frontiers in Human Neuroscience
- 1
These authors contributed equally to this manuscript.