Back

D2-expressing neurons of the anterior paraventricular nucleus of the thalamus as key players for modulating relative aversive and safety learning

Miranda, M.; Karam, E.; Tuduri, P.; Passard, M.; Drahmani, M.; Vaillant, L.; Vallianatou, C.-A.; Lavigne, M.; Kremer, E. J.; Ster, J.; Perisse, E.; Trouche, S.

2025-09-18 animal behavior and cognition
10.1101/2024.12.30.630766 bioRxiv
Show abstract

Associative learning provides a means to assign appropriate value signals to experiences that animals can later use to make appropriate value-based choices. By generalizing from prior aversive experiences and comparing with a current experience, animals can learn the relative aversive ("better or worse than") value or relative positive safety value between experiences. Yet, how such relative value signals are acquired during learning remain poorly known. The paraventricular nucleus of the thalamus (PVT) is critical for the control of salience and valence signals. It is therefore well-placed to be involved in learning relative aversive value and learned safety. Using a newly-developed conditioned place preference task, we uncover specific behavioral signatures associated with learning relative aversive value and learned safety, and related value-based choices. We then show that neurons expressing dopamine D2 receptor (D2+) in the anterior PVT (aPVT) are preferentially recruited by both relative aversive value learning and learned safety. Using a pharmacogenetic strategy, we demonstrate that aPVT D2+, but not D2-, neurons are involved in learning relative- but not absolute- aversive value. Finally, we show that aPVT D2+, but not D2-, neurons are also critical for the encoding of learned safety. Overall, our findings reveal a novel role of aPVT D2+ neurons in computing relative aversive/safety value signals between experiences during learning to promote appropriate value-based choices.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.