Dissertations, Theses, and Capstone Projects

Date of Degree

9-2026

Document Type

Doctoral Dissertation

Degree Name

Doctor of Philosophy

Program

Biology

Advisor

Ekaterina Likhtik

Committee Members

Nesha Burghardt

Andrew Delamater

Elizabeth Bauer

Robert Froemke

Subject Categories

Behavioral Neurobiology | Biology | Neuroscience and Neurobiology | Systems and Integrative Physiology | Systems Neuroscience

Keywords

Sensory-attention networks, emotion-attention networks, safety learning, sex differences, fiber photometry, electrophysiology

Abstract

Post-traumatic stress disorder (PTSD) is a trauma- and stressor-related disorder with a lifetime prevalence of 5-6% in men, and 10-12% in women (National Center for PTSD). Generalized anxiety disorder (GAD) is a highly prevalent anxiety- and depression-related psychiatric disorder with a lifetime prevalence of 1-2% in men, and 3-4% in women (National Institutes of Health). Both of these disorders disproportionately affect women. A hallmark feature of both disorders is the maladaptive generalization of fear, in which defensive responses extend beyond the original traumatic or threatening stimulus to neutral or unrelated cues. This overgeneralization can become debilitating, disrupting cognitive, emotional, and social functioning. Current clinical interventions, including exposure therapy and extinction-based approaches, aim to reduce the salience and emotional impact of fear-associated memories through repeated nonreinforced exposure to fear cues. However, extinction does not erase the original fear memory. Instead, it creates a competing inhibitory memory, leaving fear susceptible to spontaneous recovery, reinstatement, and renewal, thereby limiting long-term therapeutic efficacy.

An alternative approach involves safety conditioning, in which organisms learn that a separate cue predicts the absence of threat. Through this process, the safety cue may acquire properties of a conditioned inhibitor of fear that can compete with and suppress the expression of conditioned fear. Safety conditioning shares conceptual similarities with latent inhibition, a phenomenon in which repeated pre-exposure to a neutral cue reduces its associability during later fear conditioning. Although both conditioned inhibition and latent inhibition attenuate fear expression, they differ theoretically in their underlying associative mechanisms and attentional processes. Despite these distinctions, relatively few studies have directly compared how fear, safety, and latent inhibition cues are acquired, expressed, and processed across sexes.

The present dissertation sought to determine whether a safety cue trained through explicitly unpaired fear conditioning would function as a true conditioned inhibitor of fear and how it compared to a latent inhibition cue. To evaluate inhibitory properties, we employed the summation test, a classical behavioral assay used to determine whether a safety cue can suppress fear to an independently trained excitatory fear cue when presented in compound. According to the Rescorla–Wagner model, a true conditioned inhibitor should significantly reduce conditioned responding during compound presentations to a greater extent than a non-inhibitory control stimulus. We further investigated whether fear, safety, and latent inhibition cues are processed similarly within the emotion-attention network, or whether they engage distinct neural mechanisms across sexes.

Our findings demonstrate clear sex-dependent differences in learning and behavioral expression across fear, safety, and latent inhibition paradigms. Males and females exhibited differential acquisition and expression of defensive freezing behavior across cue types, suggesting that biological sex significantly influences inhibitory learning processes. Interestingly, neither the safety cue nor the latent inhibition cue satisfied the classical criteria of the summation test predicted by the Rescorla–Wagner framework. Nevertheless, both cues robustly suppressed fear responding in the presence of a conditioned fear cue, indicating that these inhibitory stimuli engage mechanisms capable of modulating fear expression despite failing to meet traditional associative definitions of conditioned inhibition. Additionally, analyses of estrous cycle phase revealed no significant relationship between cycle stage and defensive freezing behavior during summation testing in females.

To further characterize the neural substrates underlying these behaviors, we examined cholinergic activity in the BLA during the summation test and physiological activity within the infralimbic cortex (IL), basal forebrain (BF), basolateral amygdala (BLA), and primary auditory cortex (A1), collectively conceptualized as an emotion-attention network. Across both males and females, neural dynamics within this network revealed clear state-dependent switching between fear and safety representations, providing evidence that these circuits flexibly encode competing emotional states. Together, these findings advance our understanding of inhibitory learning, sex differences in fear regulation, and the neural circuitry underlying fear suppression, with important implications for improving therapeutic strategies targeting maladaptive fear and anxiety disorders.

Download

Share

COinS