Skip to main content
eScholarship
Open Access Publications from the University of California

UC Riverside

UC Riverside Electronic Theses and Dissertations bannerUC Riverside

Reinforcement Learning Influences Widespread Changes in Cortical Representations in a Selective Whisker Detection Task in Mice

Creative Commons 'BY' version 4.0 license
Abstract

A fundamental aspect of goal-directed behavior involves the capacity to selectively respond to specific stimuli during the decision-making process. My dissertation project is dedicated to uncovering the associations between changes in behavioral outcomes and the broader patterns of cortical activity as mice acquire proficiency in the selective detection task. In our research, we aim to delve into the neural mechanisms that underlie sensory selection (sensory detection and impulse control), employing the Widefield Calcium imaging technique. To achieve this, we conducted a training regimen with mice, employing a whisker-based selective detection paradigm where they learned to respond to preferred target stimuli while disregarding non-preferred distractor stimuli throughout the learning process. Notably, mice that achieved expertise in the task demonstrated a clear attenuation of sensory-to-motor signal propagation in distractor-aligned cortical regions. Additionally, we explored the impact of prestimulus activity in the neocortex on stimulus detection. We observed that reduced prestimulus activity in the dorsal cortex correlated with improved stimulus detection, predicting whether a response would occur or not, and resulting in faster reaction times. Finally, we investigated whether learning the selective detection task induces widespread changes across the cortex, examining whether alterations in specific behavioral measures can be linked to distinct cortical modulations. Our results showed that the learning process entails extensive neocortical adaptations as mice advance to expert-level performance in the task. This research offers valuable insights into the learning mechanisms involved in the selection process, with potential applications for understanding impairments in learning trajectories observed in certain mental health disorders.

Main Content
For improved accessibility of PDF content, download the file to your device.
Current View