No abstract

We compared procedural learning, translation of procedural knowledge into declarative knowledge, and use of declarative knowledge in age-matched normal volunteers (n = 30), patients with Parkinson's disease (n = 20), and patients with cerebellar degeneration (n = 15) by using a serial reaction time task. Patients with Parkinson's disease achieved procedural knowledge and used declarative knowledge of the task to improve performance, but they required a larger number of repetitions of the task to translate procedural knowledge into declarative knowledge. Patients with cerebellar degeneration did not show performance improvement due to procedural learning, failed to achieve declarative knowledge, and showed limited use of declarative knowledge of the task to improve their performance. Both basal ganglia and cerebellum are involved in procedural learning, but their roles are different. The normal influence of the basal ganglia on the prefrontal cortex may be required for timely access of information to and from the working memory buffer, while the cerebellum may index and order events in the time domain and be therefore essential for any cognitive functions involving sequences.

We compared the performance of 12 patients with cerebellar atrophy (CA) and 12 normal controls matched for age and education on the Tower of Hanoi, a nine-problem task that requires cognitive planning. CA patients performed significantly worse than controls on this task despite no difference in planning and between-move pause times. A reanalysis of the data using just the subgroup of patients with pure cerebellar cortical atrophy (CCA) (N = 9) replicated the above results and also showed that CCA patients had significantly increased planning times compared with controls. Neither age, sex, education level, severity of dementia, word fluency, response time, memory, nor visuomotor procedural learning predicted CA or CCA performance. This deficit in cognitive planning suggests a functional link between the cerebellum, basal ganglia, and the frontal lobe concerning specific cognitive processes. However, the exact role of the cerebellum in cognitive planning remains undetermined.

There is evidence from animal experiments that the cerebellum and its associated brainstem circuitry are involved in the acquisition of the conditioned response. In order to obtain evidence for their involvement in humans, we studied classical delay conditioning, using the eye-blink conditioned response, in five patients with pure cerebellar cortical atrophy and seven patients with olivopontocerebellar atrophy. The results were compared with those obtained in a group of neurologically healthy volunteers matched with the patients for age and sex. The two groups of patients had similar abnormalities in the acquisition of the conditioned response and produced fewer conditioned responses than in the control subjects in any given block of trials. Many of the patients' conditioned responses were inappropriately timed with respect to the conditioned stimulus. These results support the role of the cerebellum in the expression and timing of the conditioned response.