Dissertations, Theses, and Capstone Projects

Date of Degree

9-2025

Document Type

Master's Thesis

Degree Name

Master of Science

Program

Cognitive Neuroscience

Advisor

Jeff Beeler

Committee Members

David Johnson

Subject Categories

Cognitive Neuroscience

Keywords

anorexia nervosa, activity-based anorexia, vulnerability, hyperactivity, eating disorders

Abstract

Anorexia nervosa (AN) is a severe psychiatric disorder with the highest mortality rate among mental health conditions, yet effective biological treatments remain limited. The activity-based anorexia (ABA) paradigm, which combines food restriction with wheel access to model the hyperactivity and self-starvation cycle characteristic of AN, shows considerable individual variability in vulnerability. Understanding what predisposes certain individuals to develop the full ABA phenotype while others remain resilient could inform early intervention strategies for human AN. This study employed continuous behavioral monitoring technology to identify early predictors of vulnerability to activity-based anorexia in mice, with the goal of discovering behavioral biomarkers that precede the onset of severe weight loss. Fifteen C57BL/6 mice (8 males, 7 females, aged 74-104 days) underwent a four-phase experimental protocol: habituation, FED training, seven-day baseline with ad libitum feeding, and food restriction to a 2-hour daily window. Wheel running activity and feeding behavior were continuously monitored using rotary encoders and Feeding Experimental Device (FED3) technology. Animals were removed when body weight dropped below 75% of baseline. Correlation analyses examined relationships between baseline and early restriction behaviors with survival time (days to 25% weight loss). All mice reached the termination criterion within 2-10 days (mean = 5.3 ± 2.2 days). Locomotor activity emerged as the strongest predictor of vulnerability across multiple temporal windows. Baseline wheel running activity during ad libitum feeding showed a robust negative correlation with survival (r = -0.834, 95% CI [-0.94, -0.56], p < 0.001), indicating that hyperactivity predisposition exists before food restriction onset. This relationship strengthened during early food restriction (r = -0.881, p < 0.001), with vulnerable mice escalating from ~8,000 to >15,000 revolutions/day within three days. Microstructural analysis revealed that longer baseline running bout duration also predicted reduced survival (r = -0.583, p = 0.023), suggesting that sustained running patterns contribute to vulnerability. Age showed a significant positive correlation with survival (r = 0.532, p = 0.041), confirming heightened adolescent vulnerability. Feeding behavior metrics showed weaker and largely non-significant associations with survival outcomes. Vulnerability to activity-based anorexia can be predicted from pre-existing behavioral patterns, particularly baseline hyperactivity and sustained running bout structure. The stronger predictive value of locomotor activity compared to feeding measures suggests that dysregulated reward processing and activity regulation, rather than primary appetite dysfunction, may be fundamental to AN vulnerability. These findings provide compelling evidence for objective, measurable predictors of eating disorder susceptibility and support the development of early identification strategies targeting high-activity individuals who begin dietary restriction. The results have important translational implications for understanding AN etiology and developing prevention approaches in at-risk populations.

Download

Share

COinS