An edge-centric approach to discerning the neural networks underlying event script processing

Poster Session C - Sunday, April 14, 2024, 5:00 – 7:00 pm EDT, Sheraton Hall ABC

Yongzhen Xie¹ (yongzhen.xie@mail.utoronto.ca), Alexander Barnett¹; ¹Department of Psychology, University of Toronto

Every day, we rely on our schematic event scripts to interpret and predict the flow of our experiences. Many brain regions have been shown to contribute to event script processing, but the underlying functional networks have not yet been disentangled. Here, we aimed to discern the neural networks that engage in script processing by examining how deviations from real-world scripts alter functional connectivity in the brain. Leveraging an edge-centric functional connectivity approach, we clustered pairs of brain regions into communities associated with video viewing. These communities encompassed regions from the hippocampus, the default mode network (DMN), and the ventral visual stream (VVS). We then inspected the connectivity of each community using the fMRI data from Baldassano et al. (2018), in which individuals viewed videos that either adhered to naturalistic event scripts (intact condition) or deviated from these scripts (scrambled condition). We found that communities involving the hippocampus, DMN, and VVS showed higher connectivity at video event boundaries in the scrambled condition compared to the intact condition, which suggests that these communities are responsive to the adherence to real-world scripts. Moreover, as individuals’ exposure to scrambled videos increased, the difference in functional connectivity between intact and scrambled conditions diminished, particularly in neural communities involving DMN. This finding indicates that the functional connectivity of DMN can adapt to deviations from event scripts through repeated exposure. In conclusion, our results reveal flexible functional networks that not only engage in real-world script processing but also adapt to experiences diverging from script schemas.

Topic Area: LONG-TERM MEMORY: Semantic