Testing a unified spectral temporal context model

California Polytechnic State University Foundation, San Luis Obispo CA

Investigators

Abstract

Memory is essential for human life, supporting adaptive behavior and core aspects of personal identity. This research project aims to understand more about episodic memory - or memory for specific events or episodes in our lives - and the central role that time plays in it. How do we recall significant life-changing events such as our graduation from high school, or our wedding day, or even ordinary, daily events such as where we parked our car this morning? This research probes how the amount of elapsed time between the occurrence of the event and our recollection of that event affects myriad aspects of our memory. How long ago the event occurred affects multiple aspects of memory function, such as the quality and availability of the memory or whether other memories interfere with the memory of the event we are trying to recall. Another key question is whether memory can be enhanced with additional cues and information. The goal of this research is to develop a unified theoretical model of memory, that accounts for the many effects of time in shaping our memories. In order to develop the best explanatory and predictive model of memory, this project aims to create and test such an account in the form of a biologically realistic computational model that incorporates recent findings on the neurophysiology of memory systems. An essential aspect to these neural findings is that representations underlying memory - which might be said to form a temporal context - drift over a wide spectrum of time scales, on the order of seconds, minutes, hours, or even days to weeks to months and years. The computational model simulates and replicates these behavioral findings that reveal how temporal context affects episodic memory. An additional goal is to use empirical data from neural recordings over unusually long-time scales to empirically test the extent of dynamic neural drift of memory representations in humans. The model incorporates neural and behavioral data that sheds light on the phenomena that we can often remember the gist of an episode, but over time will forget some of the details of that episode. Finally, it has been shown that representations not only drift, but also suddenly shift at boundaries between events and spatial environments. This project expands our efforts, leading to a more dynamic understanding of the consequences of these sudden shifts both behaviorally and in neural data. In this way, it is possible to achieve a deeper understanding of memory, and how it changes at a behavioral and neural level over time. More broadly, this project also aims to strongly increase the participation of students from historically underserved groups in STEM fields. Knowledge generated from this research is incorporated into courses and workshops at the PIs’ university and made freely available for a broader public audience. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.