Remembering the future: Interactions between sensation, memory, and behavior

01.09.2018 bis 28.02.2023

Memory is typically treated as a representation of the past, including salient stimuli and events. A key purpose for the retention of these memories, though, is to provide predictive information that affects the animal’s future behavior - where it will go and what it will do. However, as the world is rarely static, memory requires continuous updating from sensation. While sensation and memory have been well studied separately, little is known about how they interact. This proposal aims to understand how behavior, sensation, and memory predict each other and how these predictions are altered based on an animal’s experience and the predictability of its environment. We hypothesize that memory evolves from being something that is learnt from (i.e. driven by) sensation to being used to predict (i.e. drive) sensation and behavior.

We will test this hypothesis by designing an environment with static (predictable) and mobile (unpredictable) features. Housing mice in this environment, we will combine long-term recordings from large populations of neurons in the visual cortex and the hippocampus, high-resolution behavioral measurements, and statistical modeling of neural dynamics and behavior. Together, these efforts will provide insight into how we use representations of the past to inform our future.