Distinct prelimbic cortex ensembles encode response execution and inhibition.

Rajtarun Madangopal, Ph.D.

Featured Paper of the Month – January 2026

Published in PNAS by Rajtarun Madangopal, Ph.D. and Bruce Hope, Ph.D. of the NIDA IRP Neuronal Ensembles in Drug Addiction Section.

Summary

Learning when to initiate or withhold actions is essential for survival, requiring the integration of past experiences with new information to adapt to changing environments. In this study, NIDA IRP researchers used single-cell calcium imaging to longitudinally track hundreds of brain cells (neurons) in rats across three phases: when they pressed a lever for food rewards (Training), as they learned to stop pressing when the reward was removed (Extinction), and when they resumed pressing after a small, non-contingent reward was reintroduced (Reinstatement). They applied computational tools to predict behavior from brain activity and applied an in silico deletion approach to simulate the effects of removing specific neurons. Their analyses showed that distinct and non-overlapping populations – called ensembles – were active during Training and Extinction, to support response execution and inhibition, respectively. These findings highlight ensemble-based encoding of multiple, even opposing, learned associations within the same region, demonstrating how selective ensemble recruitment enables behavioral flexibility under changing contingencies. Further, when the behavior returned during Reinstatement, these same ensembles were re-engaged, indicating a potential neural basis for relapse.

Publication Information

Madangopal, Rajtarun; Zhao, Yuan; Heins, Conor; Zhou, Jingfeng; Liang, Bo; Barbera, Giovanni; Lam, Ka Chun; Komer, Lauren E; Weber, Sophia J; Thompson, Drake J; Gera, Yugantar; Pham, Diana Q; Savell, Katherine E; Warren, Brandon L; Caprioli, Daniele; Venniro, Marco; Bossert, Jennifer M; Ramsey, Leslie A; Jedema, Hank P; Schoenbaum, Geoffrey; Lin, Da-Ting; Shaham, Yavin; Pereira, Francisco; Hope, Bruce T

Distinct prelimbic cortex ensembles encode response execution and inhibition Journal Article

In: Proc Natl Acad Sci U S A, vol. 122, no. 37, pp. e2505378122, 2025, ISSN: 1091-6490.

Abstract | Links

@article{pmid40920924,

title = {Distinct prelimbic cortex ensembles encode response execution and inhibition},

author = {Rajtarun Madangopal and Yuan Zhao and Conor Heins and Jingfeng Zhou and Bo Liang and Giovanni Barbera and Ka Chun Lam and Lauren E Komer and Sophia J Weber and Drake J Thompson and Yugantar Gera and Diana Q Pham and Katherine E Savell and Brandon L Warren and Daniele Caprioli and Marco Venniro and Jennifer M Bossert and Leslie A Ramsey and Hank P Jedema and Geoffrey Schoenbaum and Da-Ting Lin and Yavin Shaham and Francisco Pereira and Bruce T Hope},

url = {https://pubmed.ncbi.nlm.nih.gov/40920924/},

doi = {10.1073/pnas.2505378122},

issn = {1091-6490},

year  = {2025},

date = {2025-09-01},

urldate = {2025-09-01},

journal = {Proc Natl Acad Sci U S A},

volume = {122},

number = {37},

pages = {e2505378122},

abstract = {Learning when to initiate or withhold actions is essential for survival, requiring the integration of past experiences with new information to adapt to changing environments. The prelimbic cortex (PL) plays a central role in this process, with a stable PL neuronal population (ensemble) recruited during operant reward learning to encode response execution. However, it is unknown how this established reward-learning ensemble adapts to changing reward contingencies, such as reward omission during extinction. Specifically, does the same ensemble adjust its activity to support behavior suppression, or is a distinct ensemble recruited for this new learning? Our data reveal that operant extinction learning recruits a distinct PL Extinction ensemble to support response inhibition, and concerted engagement of both ensembles encodes both ongoing and subsequent context-specific behavior. Using single-cell calcium imaging, we longitudinally tracked PL neurons in rats as they pressed a lever for food rewards (Training), learned to suppress responding upon reward omission (Extinction), and reinstated responding following a noncontingent "priming" pellet (Reinstatement). We trained decoders on individual rats' PL activity patterns to predict trial-wise responses and used an in silico deletion approach to identify separate PL Training and Extinction ensembles associated with response execution and inhibition, respectively. Critically, both ensembles were reengaged and maintained their distinct roles during Reinstatement. These findings highlight ensemble-based encoding of multiple, even opposing, learned associations within the same region, demonstrating how selective ensemble recruitment enables behavioral flexibility under changing contingencies.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}