About

Patricia Janak, Ph.D., is a Bloomberg Distinguished Professor at Johns Hopkins University with joint appointments in the Department of Psychological and Brain Sciences in the Krieger School of Arts and Sciences and in the Solomon H. Snyder Department of Neuroscience in the Johns Hopkins School of Medicine. She received her BA in psychology and biology from Rutgers College and earned her Ph.D. in biological psychology from the University of California, Berkeley. Following postdoctoral positions at the Wake Forest School of Medicine and the National Institute on Drug Abuse, she joined the Ernest Gallo Clinic and Research Center at the University of California, San Francisco, as an Assistant Professor in 1999. She was promoted to Associate Professor in 2006 and was named the Howard J. Weinberger Endowed Chair in Addiction Research at UCSF in 2011. In 2014, she transitioned to Johns Hopkins University to continue her research and academic career. Her work focuses on addiction research, and she leads the Janak Lab, which investigates neural mechanisms underlying addiction and reward-related behaviors.

Research topics

• Computer Science
• Psychology
• Neuroscience
• Artificial Intelligence
• Machine Learning
• Biology
• Medicine
• Chemistry
• Internal medicine
• Biochemistry
• Social psychology

Selected publications

• Uncertainty and reward histories have distinct effects on decisions after wins and losses

  Scientific Reports · 2026-01-31 · 1 citations

  articleOpen accessSenior author

  Intelligent behavior necessitates an adaptive integration of feedback. It is well-known that animals asymmetrically learn from positive and negative feedback. While asymmetrical learning is a robust behavioral effect, the latent computations behind how animals represent their environments and use this to differentially weight wins and losses is poorly understood. Here we tested whether and how uncertainty and reward history modulate the weights placed on wins and losses using a behavioral data set collected in rats. We propose a reinforcement learning model that integrates uncertainty history via an unsigned average reward prediction error and a separate subjective reward history component. We showed that in a dynamic probabilistic reversal learning task with blocks of variable reward predictability, ongoing estimation of uncertainty history and reward history both distinctly influenced rats' sensitivity to wins and losses. In more predictable environments, and under low uncertainty levels, i.e., when rats were certain in making 'correct' choices, rats weighted wins more than losses, as indicated by a higher win-stay, and lower lose-shift probability. This asymmetrical learning strategy enabled rats to remain with the correct action, while discounting the influence of rare losses. Further, male rats were more impacted by their uncertainty history when making win-stay decisions compared to females. Hence, we found sex-specific contributions of these latent computations in modulating behavior. We overall demonstrate that asymmetrically weighting wins and losses could form an important behavioral strategy when adapting to ongoing changes in reward and uncertainty history.

  PublisherOA PDFDOI

• Sequence termination cues drive automated habit-like strategy via dopamine-mediated processes

  Neuropsychopharmacology · 2026-05-02

  article
  PublisherDOI

• Memory erasure by dopamine-gated retrospective learning

  bioRxiv (Cold Spring Harbor Laboratory) · 2026-01-21 · 1 citations

  articleOpen access

  Erasing outdated memories is crucial for adaptive behavior. Yet once a cue-outcome association is learned, repeated cue exposure without outcome suppresses conditioned behavior without erasing the underlying memory. This allows rapid behavioral recovery when outcomes are reintroduced. Here, we confirm this limitation for standard "prospective extinction" protocols that present cues without the associated outcome, but show that true memory erasure is achieved by inverting the paradigm: presenting outcomes without associated cues, i.e., "retrospective extinction". We demonstrate that orbitofrontal cortex activity at outcome is necessary for the rapid behavioral recovery following prospective extinction, and that mesolimbic dopamine activity at outcome is necessary for retrospective extinction. These findings reconceptualize extinction mechanisms and suggest complementary strategies to mitigate relapse and erase maladaptive memories.

  PublisherDOI

• Fentanyl reinforcement history has sex-specific effects on multi-step decision-making

  Neuropharmacology · 2026-01-29

  articleOpen accessSenior author
  PublisherOA PDFDOI

• The Central Nucleus of the Amygdala Encodes the Motivation to Pursue Ethanol

  bioRxiv (Cold Spring Harbor Laboratory) · 2026-05-14

  articleSenior author

  Abstract Prior studies implicate the central nucleus of the amygdala (CeA) in reward motivation, yet how this region encodes motivation for ethanol (EtOH) seeking and consumption–and how this compares to encoding of natural rewards—remains poorly understood. We recorded single-unit neural activity from rats during a cued instrumental task in which lever insertion on each trial indicates the opportunity to lever press for ethanol and post-press lever retraction signals ethanol delivery. We found neurons responsive to multiple trial events (lever insertion and retraction cues, lever press, port entry, and reward licks), including neurons with rhythmic activity entrained to licking during EtOH consumption. Notably, CeA neural responses to the lever insertion cue at trial start encoded both the likelihood and speed of engagement during reward seeking. Using a supervised classifier, we found that trial engagement and motivation level could be decoded from pre-trial and lever insertion response periods. Finally, we assessed whether these effects generalized to natural reward seeking. Sucrose-rewarded rats (both ethanol-exposed and ethanol-naïve) showed higher motivation (more rewards earned and shorter first-press latencies) and stronger CeA recruitment and responses at lever insertion than ethanol-rewarded rats, in agreement with the notion that CeA circuits amplify reward-predictive cue signals to facilitate rapid action initiation under high motivational states. Together, these findings indicate that CeA neurons dynamically encode motivational states, with tonic activity and reward-predictive cue responses predicting both engagement in reward seeking and the vigor of action initiation. We suggest these patterns of neural activity drive ethanol and sucrose pursuit. Significance Statement How motivational states drive reward-seeking is a central question in systems neuroscience, yet the mechanisms linking neural signals to action initiation remain poorly understood. Recording single-unit activity in the amygdala central nucleus (CeA) during cued ethanol self-administration revealed that neural responses to reward-predictive cues encoded both the decision to engage in reward seeking and the vigor of action initiation, and motivational state was decoded from cue-evoked signals and tonic activity even before cue presentation. Consistent with greater motivation for the more palatable sucrose reward, CeA neurons showed stronger cue responses and greater anticipatory firing preceding tongue contact with reward, compared with ethanol-rewarded rats. These findings identify CeA activity patterns that link motivational state to the rapid initiation of goal-directed action.

  PublisherDOI

• Transcriptional response to chronic long-access fentanyl self-administration in rat habenula and amygdala

  bioRxiv (Cold Spring Harbor Laboratory) · 2025-11-28 · 1 citations

  preprintOpen access

  Fentanyl is a potent synthetic opioid associated with overdose. However, little is known about fentanyl-induced molecular adaptations in the habenula and amygdala, two brain regions implicated in opioid use and withdrawal. We performed bulk RNA-sequencing in the rat habenula and amygdala to identify transcriptomic changes associated with fentanyl intake. Rats self-administered intravenous saline or fentanyl over 22-24 days. Ninety minutes following the final session, brains were collected for transcriptomic profiling. In Hb, we identified 453 differentially expressed genes (DEGs) between saline and fentanyl rats, with upregulated genes associated with synaptic transmission and ionic conductance. In amygdala, we identified 3,041 fentanyl-associated DEGs with upregulated genes implicated in metabolic and vesicular functions. Downregulated genes in both regions were enriched for extracellular matrix functions. Integration of DEGs with single-cell RNA-sequencing data from rodents and humans revealed that fentanyl DEGs were enriched in specific habenula and amygdala cell type markers. Furthermore, fentanyl downregulated DEGs in amygdala were enriched in genes associated with risk for substance use disorders. Together, we define how fentanyl intake alters transcriptional programs in the rat habenula and amygdala, and we link these changes to specific human cell types and risk genes for neuropsychiatric disorders and addiction.

  PublisherDOI

• Disrupted decision making in adult male rats after prolonged withdrawal from ethanol vapor exposure

  Alcohol Clinical and Experimental Research · 2025-08-18

  articleOpen accessSenior author

  BACKGROUND: Chronic alcohol use disrupts decision making and cognitive flexibility, but its long-term effects after adult exposure remain poorly understood. While prior studies have shown moderate disruptions in behavioral flexibility weeks after ethanol (EtOH) withdrawal, we tested whether such deficits persist after prolonged abstinence (>6 months) in male rats. METHODS: Male rats underwent 4 weeks of EtOH vapor or air exposure starting at postnatal day 75. After approximately 7 months of withdrawal, rats (postnatal day ~300) completed a series of reversal learning tasks featuring distinct reward probabilities with deterministic (100/0%) and uncertain (90/10% and 80/20%) schedules. We assessed win-stay and lose-shift behaviors, as well as response latencies. Reinforcement learning (RL) models were fit hierarchically to identify group differences in value updating during reward-guided decision making. RESULTS: When rats had novel reversal experience under deterministic conditions, ethanol (EtOH)-exposed rats showed increased lose-shift behavior in trials preceding the reversal point. When a moderate outcome uncertainty (90/10) was introduced, EtOH rats displayed reduced win-stay behavior in trials after reversal when learning new contingencies. With continued training under greater outcome uncertainty (80/20), these changes were not manifested. Furthermore, RL models revealed that rats employed distinct learning processes depending on the reward schedule. Across all RL models, EtOH-exposed rats exhibited a consistently high decay of unchosen action values, either through faster forgetting or enhanced updating after negative outcomes. CONCLUSIONS: These findings demonstrate that a remote history of adult alcohol exposure can lead to persistent, uncertainty-sensitive disruptions in reward-based decision making, even after exceptionally long withdrawal.

  PublisherOA PDFDOI

• WITHDRAWN: Modeling hidden cognitive states reveals acute and chronic effects of fentanyl on decision-making

  bioRxiv (Cold Spring Harbor Laboratory) · 2025-08-05

  preprintOpen access

  The cognitive mechanisms underlying behavior are often dynamic, shifting gradually or abruptly over time scales spanning years, to weeks, to minutes. Whether drug-induced changes in learning and decision-making follow similarly dynamic patterns remains unclear. To address this, we apply a reinforcement learning model to choice data from rats performing a two-step task for oral fentanyl and sucrose rewards. The model contains a set of agents with their own learning and decision-making rules that differentially influence choice, and, critically, each agent's contribution to choice is allowed to vary across latent states that fluctuate over time. Using a dimensionality reduction method to align latent states across subjects, we identified three distinct states reflecting mixtures of goal-directed, habitual, and novelty-driven strategies. We found that acute fentanyl reward increased the frequency of transitions out of a goal-directed state into a habit-driven state, while chronic fentanyl exposure selectively diminished goal-directed control within a habit-dominant state, independent of sex. Together, these results demonstrate that fentanyl reshapes both the dynamics and cognitive architecture of decision-making, underscoring the utility of latent-state modeling combined with dimensionality reduction for uncovering drug-driven cognitive changes.

  PublisherOA PDFDOI

• Contextual cues facilitate dynamic value encoding in the mesolimbic dopamine system

  Current Biology · 2025-01-23 · 8 citations

  articleOpen access
  PublisherOA PDFDOI

• mWAKE in the Central Amygdala Regulates Fear Learning and Memory

  Journal of Neuroscience · 2025-08-20

  articleOpen access

  The central amygdala (CeA) is an important neuronal hub that integrates external sensory inputs and information about internal states to regulate a range of innate and learned behaviors, including fear learning and memory. Prior studies, leveraging robust fear conditioning assays, have delineated detailed circuit mechanisms underlying the acquisition and recall of fear memories. However, the specific molecular mechanisms underlying these processes in the CeA remain poorly understood. Here, we investigate the role of the clock output molecule mWAKE/ANKFN1 within the CeA of male mice in fear learning and memory. mWAKE is expressed in multiple neuronal subclusters in the lateral CeA. Surprisingly, mWAKE levels do not exhibit rhythmic expression in the CeA. In line with this observation, expression of the core clock genes PER2 and BMAL also does not cycle in the CeA. Consistent with prior studies, loss of mWAKE function increases intrinsic excitability of CeA neurons. Furthermore, conditional knock-out of mWAKE and chemogenetic activation of CeA mWAKE neurons impair fear learning and memory. Finally, we show that mWAKE levels in a subset of CeA neurons are reduced following fear conditioning. These findings suggest a potential molecular mechanism modulating the activity and function of CeA neurons in fear learning and memory.

  PublisherOA PDFDOI

Recent grants

• Amygdala neural circuits in alcohol intake

  NIH · $2.1M · 2019–2025

• NIH Grant R03DA024363

  NIH · $369k · 2010

• NIH Grant R01AA014925

  NIH · $3.6M · 2016

• NIH Grant R01DA015096

  NIH · $2.5M · 2011

• NIH Grant R01AA018025

  NIH · $1.7M · 2014

Frequent coauthors

• Laia Castell

  Johns Hopkins University

  126 shared

• Giuseppe Gangarossa

  Unit of Functional and Adaptive Biology

  118 shared

• Marcelo Rubinstein

  Experimental Medicine and Biology Institute

  116 shared

• Pauline Tarot

  Université de Montpellier

  116 shared

• Laura Cutando

  Inserm

  116 shared

• Ha-Rang Kim

  Université Paris Cité

  112 shared

• Valentine Le Gall

  Université de Montpellier

  112 shared

• Daniel Jercog

  Neurocentre Magendie

  112 shared