UCLA

The medial prefrontal cortex is involved in many functions, including memory, decision making, and emotional regulation. This region is strongly tied to a variety of psychiatric diseases, especially those involved in balancing approach-avoidance conflicts, such as anxiety disorders and depression. Understanding the cellular mechanisms that regulate mPFC functions is critical for learning how they become maladaptive and contribute to disease.

Threat avoidance is a critical behavior for surviving in a dynamic environment. However, one must balance threat avoidance with other behaviors, such as foraging and nesting. Many psychiatric disorders including anxiety disorders, depression, and phobias feature excessive, maladaptive avoidance as a central symptom. Understanding how regulation of avoidance behavior emerges under normal circumstances is critical for understanding how maladaptive regulation produces excessive avoidance in patients.

As new methods in neuroscience allow the examination of neural activity in ever greater resolution, there is a growing need for tools that can automate the processing and analysis of animal behavior and align it with respect to simultaneously recorded data streams. This thesis presents a new software program, BehaviorDEPOT, that allows forecasting of animal behavior from keypoint tracking and generates detailed reports of kinematic and postural behavior statistics to facilitate downstream analyses. This pipeline provides a new tool aimed at non-computational users that bridges estimation of an animal’s location with the extraction of detailed information about the animal’s posture, movement, location, and behavior.

This thesis combines the capabilities of BehaviorDEPOT with miniscope imaging of mPFC to examine the emergence of activity patterns associated with avoidance learning. Examining mPFC dynamics throughout rapid learning, this thesis reveals how changes in neural representations of threatening cues and safe locations are some of the first to emerge during learning. This thesis characterized signatures of learning in mPFC and identified an emergent population activity pattern during the onset of conditioned tones that correlates with learning rate and emerges before changes in behavior. This body of work contributes a useful tool that has already been adopted by neuroscientists and reveals activity patterns in mPFC that will contribute to our knowledge of maladaptive aversive learning.