Schedule of Events | Search Abstracts | Invited Symposia | Symposia | Poster Sessions | Data Blitz

Tracking Event Boundary Processing in Language: Hierarchical Dynamics of Comprehension and Memory

Poster Session D - Monday, March 9, 2026, 8:00 – 10:00 am PDT, Fairview/Kitsilano Ballroom
Also presenting in Data Blitz Session 1 - Saturday, March 7, 2026, 10:30 am – 12:00 pm PST, Salon ABC.

Mar Domínguez-Orfila^1,2 (), Ruth De Diego-Balaguer^1,2,3, Lluís Fuentemilla^1,2; ¹Institute of Neuroscience, University of Barcelona, ²Bellvitge Institute for Biomedical Research, ³nstitució Catalana de Recerca i Estudis Avançats (ICREA)

Understanding a narrative requires constructing a coherent situation model by segmenting information into discrete events, typically marked by shifts in time, space, or goals. While most research on event segmentation relies on continuous audiovisual stimuli where event boundaries must be inferred, language provides inherently discrete boundaries, yet fewer studies have examined how this process unfolds during the comprehension of written narratives. Most language-based studies focus on behavioral measures, providing limited insight into the real-time neural dynamics underlying event model updating. Here, we combine word-by-word reading times (RTs) with scalp EEG to examine how readers update event representations and how these dynamics influence memory organization. We used narratives with two hierarchical levels: stimuli adapted from Ding & Zacks (2025) comprised ten short stories with a nested structure of fine- and coarse-level events. We hypothesize that each level exhibits distinct behavioral and neural signatures supporting event updating and episodic memory formation. Participants read stories word by word in a self-paced task while EEG was recorded, followed by a distractor and a free recall task. RTs increased at both boundary types, with coarse boundaries showing a post-boundary slowdown relative to fine boundaries, indicating more effortful integration. EEG data mirrored these behavioral effects, revealing a larger late positivity for coarse than for fine boundaries. Finally, fine-level events were more likely to be recalled when reading times were higher at boundary positions, but not coarse-level events. These results emphasize language as a precise framework for investigating the dynamics and structure of event segmentation.

Topic Area: LONG-TERM MEMORY: Episodic