UC San Diego

This Master’s thesis attempts to assess the question of what the role of the retrosplenial cortex (RSC) is in foraging task - a task in which a mouse is supposed to make a value-based decision that uses the internal representation of value, rather than an external sensory cue or perceptual input. This paper also attempts to find out what the effect of inactivation of the surrounding areas’ projections to RSC is on the same task. To test these questions, three experiments have been conducted - Experiment I utilized both optogenetic inactivation and NMDA lesions of RSC, and the other two experiments examined the effects of inactivation of the projections from RSC, ACC, and V1 to RSC (Experiment II) and projections from CA1 and subiculum to RSC (Experiment III).

Collected data were analyzed in three ways: (1) basic behavioral metrics such as reward rate and choice probability; (2) logistic regression behavioral model that estimates how the mouse integrates recent experience to make a decision on the upcoming trials; (3) fitting behavior data to a reinforcement learning algorithm which models the internal decision making process. The first two experiments have been completed - showing that acute optogenetic inactivation of RSC selectively impaired the reward-history-based strategy, while also revealing involvement of ACC projections to RSC in this cognitive process. Experiment III has not been completed due to the Covid-19 pandemic.