Name: Aleah Davis

Masters Thesis Proposal Meeting

Date: Friday, February 20th, 2026

Time: 10:00 am - 11:00am

Location: Virtual, Meeting link click here

Thesis Chair/Advisor:

Eric Schumacher, Ph.D. (Georgia Tech)

Thesis Committee Members:

Shella Keilholz, Ph.D. (Emory)

Mark Wheeler, Ph.D. (Georgia Tech)

Title: Functional connectivity in dynamic brain networks across different levels of suspense during naturalistic viewing
 

Abstract: Attention is a limited cognitive resource that is dynamically allocated based on task demands, motivation, and stimulus salience. Contemporary neuroscience characterizes attention as emerging from coordinated interactions among large scale brain networks, including the dorsal attention network (DAN), ventral attention network (VAN), frontoparietal control network (FPCN), and default mode network (DMN). One way to measure fluctuations in attentional engagement in these networks is to investigate how attention changes during the presentation of narrative content (e.g., movies, stories, etc.).  The complex audiovisual content of films mirror content in real-world settings as people adjust their attention to complex stimuli in their environment. Suspenseful narratives reliably heighten engagement, narrow attentional focus, reduce mind wandering, and enhance memory for central plot elements. Prior neuroimaging studies show increased activation in attentional networks and suppression of the DMN during moments of high suspense, but these findings rely largely on static measures of brain activity and do not capture how network interactions dynamically evolve over time.

The central question of this study is whether narrative suspense drives reconfiguration of functional brain networks toward stable, high engagement states that resemble sustained attention. Although dynamic connectivity has been linked to fluctuations in attention during simple tasks, its role in naturalistic narrative engagement has not been investigated. This project addresses this gap by applying quasi-periodic pattern (QPP) analysis to existing fMRI film viewing datasets with the goal of integrating continuous suspense ratings with time varying functional connectivity to identify recurring patterns of network coordination during high and low suspense.

Movie clips will be classified by average suspense and analyzed using a sliding window QPP algorithm to quantify connectivity among attentional networks. It is expected that high suspense periods will show stronger DAN/FPCN coupling, reduced DMN activity, and more frequent and stable “in-the-zone” network configurations. These findings would provide evidence that narrative engagement reflects dynamic network reorganization and would advance network based models of attention in real world contexts.