Abstract
SummaryThe dorsal and ventral regions of the hippocampus (dCA1 and vCA1) are critical for contextual fear conditioning tasks, yet the dynamics of hippocampal neuronal representations during memory acquisition, retrieval, and extinction processes have yet to be fully elucidated. The canonical theory is that the hippocampus generates and retrieves spatial maps of contexts during memory acquisition and retrieval. It is hypothesized the hippocampus prevents memory interference by generating context representations that are dissimilar and the prediction follows that representation dissimilarity facilitates discrimination between contexts. Here, we developed a context fear memory retrieval task and combined it with 1-photon neuronal imaging in dCA1 and vCA1 of freely behaving mice to test this prediction. We identified several phenomena specific to vCA1. First, fear conditioning induced an immediate and strong representational change that was predictive of freezing behavior. During context discrimination, vCA1 representations of the threatening and neutral contexts became more similar. Third, during threatening context memory retrieval, vCA1 expressed rapid and strong context representations. These unexpected results suggest that representational similarity in vCA1 facilitates faster and more efficient network state transitions. In further support of this view, both phenomena of representational similarity and rapid context recall were no longer observed in vCA1 once fear behavior was extinguished. Together, these results reveal that vCA1 unexpectedly generates similar population codes, promoting faster transitions between network states essential for contextual fear memory retrieval.HighlightsWe established a novel context teleportation task that allows population level hippocampal recordings during key moments of threat memory acquisition, retrieval, and extinction.The ventral region of CA1 (vCA1) exhibited the greatest change in neural representation during fear memory acquisition, compared to dorsal CA1 (dCA1).After fear conditioning, representations for threatening and neutral contexts were more similar in vCA1 compared to dCA1, yet representational similarity in vCA1 supported rapid context memory retrieval.Context fear memory extinction reversed the context representation in vCA1 to patterns observed prior to fear conditioning.Graphical Abstract
Publisher
Cold Spring Harbor Laboratory