Peter Dayan: Expected and unexpected uncertainty in a hippocampal code

Abstract

Natural environments are replete with different sorts of variability, and hence different sorts of uncertainties. However, how these are represented in the brain is the subject of much debate. Here, we examine the effect of continual (expected) and surprising (unexpected) changes in the location of reward on one of the best-understood neural codes, namely hippocampal place cells. We performed calcium imaging in area CA1 in a virtual reality task in which mice locomote along a linear track and lick for a water reward whose precise location on any run may be variable within bounds (expected uncertainty), and which might also translate without warning (unexpected uncertainty).

We examine place fields in three different reference frames: one tied to the environment; a second to the location of the reward on a run; and a third in which the metric is warped by the location of the reward.  We show that high versus low expected uncertainty leads to different proportions of cells that are tied to these reference frames, and that then translate differently in the face of an unexpected change. We consider these results as a form of generalized remapping.

This is joint work with Charline Tessereau, Feng Xuan, Dan Dombeck and Jack Mellor.

Duration of the talk:

Approx. 60 minutes, followed by a discussion, with snacks and drinks.

Where

IMBIT NEXUS Lab, Georges-Köhler-Allee 201, 79110 Freiburg

More about the speaker and his research

Peter Dayan

Host

Ilka Diester

More details at

https://www.brainlinks-braintools.uni-freiburg.de/de/termine/e-id/2025/04/02/event/ical/635/