PB 8 The effect of cerebellar degeneration and Parkinson's disease on motor learning networks. Issue 10 (October 2017)

• Record Type: Journal Article
• Authors: Tzvi-Minker, E.; Bey, R.; Zimmermann, C.; Nitschke, M.; Münte, T.; Krämer, U.
• Is Part Of: Clinical neurophysiology. Volume 128:Issue 10(2017:Oct.)
• Page Start: e317
• Publication Date: 2017-10
• Subjects: Neurophysiology -- Periodicals; Electroencephalography -- Periodicals; Electromyography -- Periodicals; Neurology -- Periodicals; 612.8
• DOI: 10.1016/j.clinph.2017.06.064
• Languages: English
• ISSNs: 1388-2457

Abstract

Abstract : Background: Acquiring and retaining novel motor skills plays a fundamental role in human behavior. Theoretical models suggest that distinct cortico-striatal and cortico-cerebellar loops mediate motor learning. Parkinson's disease (PD) and spino-cerebellar ataxia (SCA) are neurodegenerative disorders which affect the basal ganglia and the cerebellum, respectively. These disorders can serve as models of how impairments in specific nodes of above-mentioned cortico-striato-cerebellar network affect network interactions and motor skill learning. The goal of this study was to investigate the critical impact of basal ganglia and cerebellar nodes on whole-brain activation patterns as well as network interactions during motor sequence learning. Based on previous studies in PD patients and cerebellar deformations patients, we hypothesized that dysfunction of cerebellum would have a global effect on the cortico-striatal-cerebellar network. By contrast, PD patients who show learning effects, will show compensatory effects in this network. Methods: FMRI was measured in a group of 13 SCA patients (SARA score > 17), 12 PD patients (UPDRS < 40), and age-matched (to each group) controls while they performed the serial reaction time task. In this task, in a hidden 12-element complex sequence of movements is to be learned implicitly. The task was comprised of sequence blocks in which participants could learn the underlying pattern, interleaved with random blocks in which no pattern of finger movements was underlying the responses. Dynamic causal modelling (DCM) was used to investigate causal network interactions. Results: As hypothesized, patients with SCA did not show behavioral evidence for learning whereas learning was evident in PD patients and controls. We found significant learning-related increase in activation in parahippocampal cortex (PHC) and cerebellar crus I when comparing SCA patients to controls. The cerebellar activity increase was observed in the non-learners of the SCA group (determined by reaction time difference between conditions), whereas learners showed activity decrease. When comparing PD to controls, we found a similar effect in PHC as well as in the substantia nigra. Comparing the two patient groups we found that learning activated bilateral caudate nucleus as well as inferior parietal lobule in SCA compared to PD, despite the lack of behavioral evidence for learning in the SCA group. DCM analysis of the cortico-striato network revealed that connection from SMA to M1 was negatively modulated by learning in controls and the learners of the SCA group, whereas the non-leaners showed a reverse pattern. Conclusions: These results suggest that both degeneration in cerebellum (SCA) and degeneration in the basal ganglia (PD) affect remote brain regions, such as the parahippocampal cortex related to learning and memory processes. The correlation of activity in cerebellar crus I with behavioral measures in the SCA group suggests a pathological increase in activity which prevents learning. Bilateral activation of caudate nucleus in the SCA group suggests recruitment of basal ganglia territories for learning, perhaps as compensation for cerebellar degeneration. Finally, we present first evidence for the influence of cerebellar degeneration on causal cortico-cortical interactions relevant for motor learning.