Hormonal Contraceptive Use Predicts Increased Salience Network Activity During Face Detection

Poster Session D - Monday, March 31, 2025, 8:00 – 10:00 am EDT, Back Bay Ballroom/Republic Ballroom

Luna Malloy¹ (luna.malloy@austin.utexas.edu), Paige Broski¹, Hamza Suhail¹, Ali Arain¹, Elizabeth Bauer¹, John Leri¹, Josh Cisler¹; ¹The University of Texas at Austin Dell Medical School

Research has demonstrated a positive association between ovarian hormones and face and facial emotion detection. The present study investigated the impact of hormonal contraceptive (HC) use on neural activation patterns, utilizing support vector machine (SVM) classifiers to predict neural activity during an fMRI face detection task. We hypothesized that HC use would predict increased activity in neural networks associated with face and/or facial emotion detection. Female participants aged 21-50 (N=89; 39 using HCs and 50 naturally cycling) completed an fMRI task in which they were asked to discern between neutral and fearful faces displayed for varying “covert” (33ms) and “overt” (500ms) durations. First, we employed SVM classifiers to build a model that predicted face vs. no face stimuli viewing based on salience network activity. Then, we used the model to generate individualized face predictions (IFPs) for each participant through a leave-one-out cross-validation approach. Finally, we employed linear mixed-effects models to explore the effects of HC use, valence (fearful/neutral), and duration (covert/overt) on IFPs. There was a main effect of HC use such that compared to naturally cycling women, women using HCs demonstrated greater salience network-based IFPs regardless of valence and duration, t(2635)=2.483, p=.013. This finding indicates that HC use may be associated with increased salience network activity when detecting faces, adding to the body of research suggesting that HC use might impact the neurocircuitry underlying face detection.

Topic Area: EMOTION & SOCIAL: Person perception