Discussion

In this chapter, we have sought to evaluate the contribution of the basal ganglia and cerebellum to temporal processing, focusing on behaviors that require precise timing in the range of hundreds of milliseconds. As shown in our review of the existing literature, it has been difficult to dissociate the functions of these regions based on patient and neuroimaging studies. While cerebellar damage is consistently linked to deficits on both time production and time perception tasks, similar deficits are reported in some studies involving patients with Parkinson's disease.

At least three issues should be kept in mind when evaluating this state of affairs. First, the inferential nature of science is strengthened by considering results from various task domains and methodologies. Our evaluation should encompass a broad range of behavioral tasks and should also include computational, anatomical, pharmacological, and physiological evidence. Ivry (1997) has argued that the case for a cerebellar timing system provides a parsimonious account of the functions of this structure in a wide range of tasks, including many in which the demands on precise timing are more subtle than in tapping or time perception tasks. Computational models based on a detailed analysis of the architecture and physiology of the cerebellar cortex are also consistent with a specialized role of this structure in representing the temporal relationships between successive events (Fiala et al., 1996; Medina et al., 2000). The case for a basal ganglia role in internal timing has not been developed to the same extent. With the exception of the PD studies reviewed above, it has been primarily based on pharmacological and lesion studies in rats, and for the most part, this work has been conducted on tasks involving intervals that span many seconds (reviewed by Meck, 1996).

Second, there are limitations in inferring basal ganglia function from studies solely involving PD patients. While this degenerative disorder clearly produces a characteristic change in basal ganglia function, the loss of dopaminergic cells also has direct and indirect effects on other neural regions, including the frontal lobes. We have recently begun testing patients with unilateral basal ganglia lesions on time production tasks (Aparicio et al., 2002), and our preliminary results suggest that their performance is normal in repetitive tapping. It is possible, however, that this group, while seemingly better matched for comparison to patients with unilateral cerebellar lesions, will fail to provide insight into basal ganglia function due to recovery and reorganization following unilateral basal ganglia damage.

Third, many neuropsychological studies have been limited to either PD or cerebellar patients. There have been few efforts to directly compare the two groups of patients within the same experiment (but see Ivry and Keele, 1989). Such comparisons offer the best opportunity to test specific hypotheses concerning the differential contributions of neural structures, which collectively are recruited in the performance of specific tasks (Casini and Ivry, 1999; Mangels et al., 1998). In this way a mapping may be established between components of a psychological process model and the underlying neural substrates. In the current study, we compared two patient groups and investigated one aspect of performance involved in paced tapping, namely, the ability to use error correction processes in order to keep responses in synchrony with a pacing signal.

0 0

Post a comment