Back

Narrative 'twist' shifts within-individual neural representations of dissociable story features

Sava-Segal, C.; Grall, C.; Finn, E. S.

2025-01-13 neuroscience
10.1101/2025.01.13.632631 bioRxiv
Show abstract

Given the same input, understanding of that input can differ depending on context. How does the brain represent the latent mental frameworks that support different interpretations of the same sensory information? In this study, participants listened to the same auditory narrative twice; the narrative had a plot twist in the middle that dramatically shifted interpretations of the story. Using a within-subject design that held both the stimulus and the individual constant, we leveraged reinterpretation-driven shifts in neural activity between the two listens to identify where representations of different narrative elements are updated under a new interpretative framework. We separated the narrative into three interrelated levels--the overall narrative model, episodes, and characters--to determine where reinterpretation-driven updates to each element were reflected in brain activity. Neural activity patterns associated with interpretations and reinterpretations of these three elements were observed in overlapping but partially distinct sets of temporal, parietal, and prefrontal regions. Results suggest that heteromodal cortex represents specific narrative elements according not to their surface features, but to latent conceptual frameworks for understanding those elements, which can shift under a new narrative interpretation. Significance statementInformation is never experienced the same way twice. Here, we investigated where and how interpretations are updated in the brain when sensory input remains identical but contextual knowledge changes how we interpret that input. We used a unique narrative stimulus with a mid-story plot twist, which participants listened to twice. We found that sets of relatively dissociable regions across the brain represent three distinct narrative elements--narrative models, episodes, and characters. These findings reveal the neural basis for how identical information is processed differently through the mental frameworks we bring to understanding it.

Published in Proceedings of the National Academy of Sciences (predicted rank #1) · training set

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.