Technical Program

Paper Detail

Paper: PS-2B.16
Session: Poster Session 2B
Location: Symphony/Overture
Session Time: Friday, September 7, 19:30 - 21:30
Presentation Time:Friday, September 7, 19:30 - 21:30
Presentation: Poster
Publication: 2018 Conference on Cognitive Computational Neuroscience, 5-8 September 2018, Philadelphia, Pennsylvania
Paper Title: Capturing the geometric structure of episodic memories for naturalistic experiences
Manuscript:  Click here to view manuscript
Authors: Andrew Heusser, Jeremy Manning, Dartmouth College, United States
Abstract: The human memory system is adept at cataloging the rich dynamics of ongoing experience. However, traditional trial-based memory experiments cannot capture these dynamics, and therefore cannot be used to study them. By constraining participants' experiences in an experiment to occur in temporally discrete trials, often arranged in a randomized order, the temporal, contextual, and emotional structure of those in-lab experiences necessarily differ from the naturalistic experiences we encounter in everyday life. Here we investigate how people verbally recall continuous videos by characterizing and relating the thematic dynamics, or ``trajectories,'' of the stimulus and participants' recalls. Unlike trial-based studies of memory wherein participants attempt to recall the precise stimuli they encounter, naturalistic recall entails capturing the fundamental geometric components of the stimulus topic trajectory. The precise words participants use to describe the stimulus, and the level of detail and number of distinct events they recall, vary considerably across participants. Nevertheless, all of the participants' recall narratives captured the fundamental ``shape'' of the original stimulus. We view this work as providing a window into which aspects of naturalistic experiences must be preserved, and which might be more flexible, in considering whether and how those experiences are remembered.