Topic: policy reversal

Performing goal-directed behaviors in an uncertain environment requires organisms to continuously evaluate the sensory consequences of their motor actions and fine-tune them to ultimately reach the goal. This evaluation involves a comparison between the predicted and actual sensory change following a movement and implementing corrective movements if there is a mismatch between the two. However, details about the neural implementation of this comparison and correction are unclear. We study this paradigm in larval zebrafish using a virtual reality setup. We first “train” the fish such that they