197. Effective Connectivity Deficits in Schizophrenia During Cerebellar-Mediated Sensorimotor Synchronization

Abstract

Background: Temporal processing and sensorimotor synchronization deficits are associated with schizophrenia spectrum disorders (SZ) and may contribute to higher order cognitive deficits (i.e. learning, memory, decision-making) characteristic of these disorders. The cerebellum is implicated in temporal processing and sensorimotor integration, and it projects to brain areas involved in cognition including parietal and frontal cortices. Therefore, behavioral and neural probes of cerebellar function can help assess contributions of the cerebellum, and its associated cortical and subcortical circuits, to cognitive processes involved in temporal processing and sensorimotor integration. It was expected that behavioral impairments in temporal processing be accompanied by a disruption in cerebellar contributions to this cerebellar-subcortical-cortical circuit in an SZ population. Methods: Using a cerebellar-dependent sensorimotor synchronization task that has previously revealed robust differences (Carroll et al., 2009; Bolbecker et al., 2014) between SZ and non-patient controls (HC), we explored differences in neural contributions to task performance in 41 HC and 30 SZ. Participants underwent 3 sessions of a 6-minute fMRI scan during a sensorimotor synchronization finger-tapping task that included tone-paced (synchronization) and self-paced (continuation) tapping at a 500 ms intertap interval (ITI). Results: Behavioral data revealed impaired timing in SZ, represented by longer (t(47) = −2.06, P < .05) and more variable (t(47) = −3.04, P < .005) ITIs during the continuation phase in the SZ group. The SZ group also tapped with a greater force (t(43.73) = −3.46, P = .001) throughout the task. Wing-Kristofferson analysis revealed more clock variance than motor variance (t(18) = 8.21, P < .001) in both groups during tapping. During continuation, primary motor (M1), cerebellum, thalamus, supplementary motor area (SMA), putamen, and inferior parietal cortex were activated (FWE P < .05). No group differences were found between SZ and HC with regards to whole-brain activation. In the SZ group compared to HC, we observed decreased effective connectivity during continuation between the cerebellum and M1 (χ2 = 7.26(1), P = .007), but not between cerebellum and thalamus or thalamus and M1. Cerebellar activation during continuation was correlated with Digit-Symbol performance (r = .49, P = .022) in the HC group only. Conclusion: Findings suggest that the cerebellum is heavily involved in the temporal component of sensorimotor continuation, particularly in the millisecond range. Moreover, findings point specifically to impairment in the cortico-cerebellar-thalamic-cortical circuit in schizophrenia, which may contribute to the observed behavioral impairments on this task.