Exploring individual differences in neural event boundaries

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

Robyn E. Wilford¹ (robyn.wilford@utoronto.ca), Erika Wharton-Shukster¹, Amy S. Finn¹, Katherine Duncan¹; ¹University of Toronto

Experience is segmented into a nested series of discrete events, separated by sharp neural state transitions that can be identified in fMRI data collected during passive narrative viewing (Baldassano et al., 2017). Current state segmentation techniques manage the noisiness of fMRI data by identifying boundaries within group-averaged data, with the logic that neural state transitions that are related to event boundaries are shared whereas noise is idiosyncratic. However, participants often disagree about the timing of event boundaries, suggesting that the perception of event boundaries is itself idiosyncratic. As such, we validated the Greedy State Boundary Search (GSBS) algorithm (Geerligs et al., 2021) for use at the individual level of analysis. We applied GSBS to individual participant data in two publicly available fMRI datasets (Alexander et al., 2017; Chen et al., 2017), subsequently averaging these individualized neural boundaries across participants to see if key results from previous work held. The resulting timeseries correlated well with normed behavioural boundary timing in key regions such as the posterior parietal cortex (p<.0001), demonstrating that GSBS can be used to identify meaningful individualized neural event boundaries. We then leveraged these individualized boundaries to explore individual differences. For example, we found that the number of boundaries increased with age (p<.0001), mirroring developmental patterns found in boundary judgements (e.g., Ren et al., 2021). These results highlight the importance of developing and validating fMRI tools for the individual level of analysis; what meaningful insights could we be missing when we average away what makes each of us unique?

Topic Area: METHODS: Neuroimaging