Individual differences in volitional social self-administration and motivation in female mice following social stress

Abstract

A mouse presses a lever. A familiar partner enters. It will do this again tomorrow, unless the world becomes dangerous. For any animal, the decision to seek social connection is not fixed; it bends under the weight of experience, and nowhere more dramatically than in the aftermath of stress. This capacity to modulate social motivation in response to threat sits at the heart of individual differences in vulnerability to neuropsychiatric illness. Most people who experience severe social stress do not develop depression or anxiety, but some do, and understanding what separates them remains one of the central unsolved problems in behavioral neuroscience. Preclinical models have long relied on involuntary social interaction tests to define stress susceptibility and resilience, collapsing a rich behavioral spectrum into a binary. What has been missing is the animal's own choice. In this thesis, I introduce a social stress operant procedure that serves as a platform that asks not whether a stressed animal avoids social contact when forced to confront it, but whether it still chooses to seek that contact when given the chance. I combine operant social self-administration, social defeat and witness defeat stress, whole-brain cFos+ density mapping, and network community structure analysis to build a multi-level account of how stress reshapes voluntary social motivation across sex. What emerges is a picture defined by divergence. Social defeat silences social seeking in males, while leaving females unchanged; witness defeat does the opposite, driving elevated motivation in females while sparing males entirely. These behavioral patterns resist binary classification, falling instead along a continuous spectrum captured by a cumulative social index. Whole-brain cFos+ density mapping and community structure analysis extend these findings beyond behavior, revealing opposite large-scale connectivity signatures for social defeat and witness defeat in female mice –two stressors, two neurobiologically distinct brain states and converging, through unbiased whole-brain analysis, on olfactory circuitry and the claustrum as key nodes linking stress experiences to prosocial behavior. Together, these findings reframe how we think about stress susceptibility. Resilience and vulnerability are not categories an animal falls into, rather they are positions along a dimension, shaped by sex, stress history, and the volitional context in which social behavior is expressed. By centering the animal’s choice, OSS opens a window onto the neurobiology of individual differences that passive procedures cannot. This work offers both a flexible behavioral platform for dissecting the mechanisms of stress-induced social dysregulation and a broader framework for thinking about how the social brain navigates threat, not as a switch that flips, but as a dial that turns. We also begin to discuss the preclinical implications that arise from this study.