Novel off-context experience constrains hippocampal representational drift.

The hippocampus forms unique neural representations for distinct experiences, supporting the formation of different memories. 1,2,3,4,5,6 Hippocampal representations gradually change over time as animals repeatedly visit the same familiar environment ("representational drift"). 7,8,9,10,11,12 Such drift has also been observed in other brain areas, such as the parietal, 13,14 visual, 15,16,17 auditory, 18,19 and olfactory 20 cortices. While the underlying mechanisms of representational drift remain unclear, a leading hypothesis suggests that it results from ongoing learning processes. 20,21,22 According to this hypothesis, because the brain uses the same neural substrates to support multiple distinct representations, learning of novel stimuli or environments leads to changes in the neuronal representation of a familiar one. If this is true, we would expect drift in a given environment to increase following new experiences in other, unrelated environments (i.e., off-context experiences). To test this hypothesis, we longitudinally recorded large populations of hippocampal neurons in mice while they repeatedly visited a familiar linear track over weeks. We introduced off-context experiences by placing mice in a novel environment for 1 h after each visit to the familiar track. Contrary to our expectations, these novel episodes decreased place cells' representational drift. Our findings are consistent with a model in which representations of distinct memories occupy different areas within the neuronal activity space, and the drift of each of them within that space is constrained by the area occupied by the others. [Display omitted] • Ca²⁺ imaging was used to quantify representational drift in a familiar environment • Mice were exposed to novel enriched environments to boost hippocampal learning • Off-context novel experiences constrain hippocampal drift rather than accelerate it • Off-context novel experiences specifically affect drift in spatial tuning A leading hypothesis suggests that representational drift in a given context stems from ongoing learning processes, such as those induced by novelty in other contexts. However, Elyasaf et al. find evidence to the contrary: novel experiences constrain hippocampal representational drift rather than accelerating it. [ABSTRACT FROM AUTHOR]