Cerebellar neurodynamics during motor planning predict decision timing and outcome on single-trial level

Abstract

The neuronal basis of goal-directed behavior requires interaction of multiple separated brain regions. How subcortical regions and their interactions with brain-wide activity are involved in action selection is less understood. We have investigated this question by developing an assay based on whole-brain volumetric calcium imaging using light-field microscopy combined with an operant-conditioning task in larval zebrafish. We find global and recurring dynamics of brain states to exhibit pre-motor bifurcations towards mutually exclusive decision outcomes which arises from a spatially distributed network. Within this network the cerebellum shows a particularly strong pre-motor activity, predictive of both the timing and outcome of behavior up to ~10 seconds before movement initiation. Furthermore, on the single-trial level, decision directions can be inferred from the difference neuroactivity between the ipsilateral and contralateral hemispheres, while the decision time can be quantitatively predicted by the rate of bihemispheric population ramping activity. Our results point towards a cognitive role of the cerebellum and its importance in motor planning. KEYWORDSCerebellum, decision making, action selection, motor planning, whole-brain calcium imaging, population ramping activity, zebrafish Recurring brain state dynamics capture representation of heat and show bifurcation for different behavioral outcomesVolumetric reconstruction of the LFM data (see also Movie S1), followed by neuronal signal extraction using our established data processing pipeline (Prevedel et al., 2014), allowed us to obtain neuronal activity traces from ~5000 of the most active neurons across the entire brain. Figure 2 shows the data for an example