ePoster

Representations of supra-second time intervals in the cerebellum of larval zebrafish

Sriram Narayanan,Aalok Varma,Vatsala Thirumalai
COSYNE 2022(2022)
Lisbon, Portugal

Conference

COSYNE 2022

Lisbon, Portugal

Resources

Authors & Affiliations

Sriram Narayanan,Aalok Varma,Vatsala Thirumalai

Abstract

Animals can discriminate time intervals ranging from microseconds to days, which serves to regulate functions as diverse as sound localization to circadian rhythmicity (Mauk and Buonomano, 2004; Buhusi and Meck, 2005). While the mechanisms that enable temporal processing at the microsecond and day scales are relatively well understood (Jeffress, 1948; Rosbash and Hall, 1989), those that enable animals to discriminate time in the behaviorally relevant seconds to minutes scale are not clear. It is believed that there might be at least two distinct interval timing mechanisms, one involving the cerebellum for the “automatic” sub-second range and another, which is independent of the cerebellum but dependent on cortico-basal ganglia circuits for the “cognitively-controlled” supra-second range (Buhusi and Meck, 2005). Given the primacy of cerebellar function in sensori-motor processing during complex behavioral tasks (Wagner et al., 2017), we sought to investigate an overlooked role for the cerebellum in sensori-motor timing at supra-second intervals. Here, we used a novel interval timing assay for larval zebrafish to investigate cerebellar involvement in timing supra-second intervals. Imaging and electrophysiology measurements reveal that when periodic optic flow stimulus is presented, cerebellar Purkinje cells show predictive activity mediated by simple spikes and climbing fiber inputs that ramp up in anticipation of flow onset. A distinct subset of Purkinje cells report prediction error signals mediated solely by climbing fiber inputs when the stimulus interval does not match expectation. These signals develop with minimal sensory evidence and can quickly adapt to new intervals if stimulus timing is altered. Robust predictive and prediction error signals are associated with faster reaction times and the modulation of behavioral latency is dependent on the cerebellum. Based on our results, we built an algorithmic model that explains how Purkinje cells mediate acquisition and updating of an internal model of supra-second stimulus time intervals.

Unique ID: cosyne-22/representations-suprasecond-time-intervals-de728829