Reset of hippocampal–prefrontal circuitry facilitates learning

Abstract

Summary Flexibly adapting to novel situations is critical for survival, which is impaired in neuropsychiatric disorders 1 . Thus, understanding whether and how novelty prepares, or primes, brain circuitry to facilitate cognitive flexibility has important translational relevance. Novelty recruits the hippocampus and medial prefrontal cortex (mPFC) 2 and may prime hippocampal-prefrontal circuitry for subsequent learning-associated plasticity. We find that novelty resets ventral hippocampal-prefrontal (vHPC-mPFC) circuitry and facilitates overcoming an established strategy. Exposing mice to novelty disrupted a previously encoded strategy by reorganizing vHPC activity to local theta (4–12 Hz) oscillations and weakening existing vHPC-mPFC connectivity. As mice subsequently adapted to a new task, vHPC neurons developed new task-associated activity, vHPC-mPFC connectivity was strengthened, and mPFC neurons updated encoding with new rules. Without novelty, mice adhered to their established strategy. Blocking dopamine D1-receptors (D1Rs) or inhibiting novelty-tagged cells expressing D1Rs in the vHPC prevented these behavioral and physiological effects of novelty. Further, D1R activation mimicked the effects of novelty. These results suggest that novelty promotes adaptive learning by D1R-mediated resetting of vHPC-mPFC circuitry, thereby enabling learning-associated circuit plasticity.

Differential encoding of place value between the dorsal and intermediate hippocampus

“…To identify the trial in which learning took place (acquisition onset) in each block, we used a state-space model, 54 which has been used in prior studies to evaluate the probability of making a correct choice in a trial-by-trial fashion (Figure 5B). [55][56][57] Rats learned to visit the sunflower-seed-baited arm reliably within approximately ten trials in both blocks 1 and 2 (Figure 5B), suggesting that the rat was capable of using the allocentric strategy. However, when the room-referenced spatial contingency for the reward location was reversed in block 3, it took $3 times as many trials for the rat to adapt to the change.…”

Complementary task representations in hippocampus and prefrontal cortex for generalising the structure of problems

“…One possibility is that, as in space, abstracted or schematic representations of tasks in cortex might be flexibly linked with the sensorimotor characteristics of a particular environment to rapidly construct concrete task representations in hippocampus, affording immediate inferences 37, 38 . Indeed, hippocampal manipulations appear particularly disruptive when new task rules must be inferred, either at the beginning of training 39 or when task contingencies change 40, 41 .…”

Coding of social odors in the hippocampal CA2 region as a substrate for social memory

“…This is likely because the default behavior of the animal prior to learning was to lick to the light cue independent of any odor so that learning the task requires the learned suppression of licking to the unrewarded odor. We observed that CA2 silencing produced a marked rightward shift of the inflection point of a logistic fit to a plot of performance to withhold licking to odor B (Park et al, 2021) (control: coefficientinflection = 3.02 days, n = 5 mice; eArch3.0: coefficientinflection = 5.26 days, n = 4 mice; p empirical = 0.04, 5000 permutations; Figure 5F). In addition, a repeated-measure ANOVA indicated a significant effect of the experimental group variable on performance with no significant interaction (Fgroup(1, 7) = 6.3, p = 0.04; Fgroup*day(5, 35) = 1.6, p = 0.2; Figure 5F).…”