Behavioral Neuroscience publications

Energy homeostasis and motivated behaviours are tightly interconnected processes coordinated by peripheral metabolic signals and central reward circuits. Among these signals, circulating triglycerides (TG) have emerged as neuromodulators of mesolimbic dopamine functions and reward processing. However, whether TG directly regulate the activity of defined neuronal populations within the nucleus accumbens (NAc) during appetitive behaviours remains unknown. Here, we investigated how central TG availability shapes the in vivo activity of NAc dopamine D2 receptor-expressing spiny projection neurons (D2R-SPNs) across distinct nutritional and metabolic states. Using in vivo fiber photometry in Drd2 -Cre mice, we monitored D2R-SPN Ca 2+ dynamics during appetitive Pavlovian conditioning and refeeding. D2R-SPNs progressively developed robust responses to a reward-predictive conditioned stimulus (CS + ) over the course of learning, while their activity during reward consumption remained stable. Acute elevation of central TG levels through carotid infusion did not alter D2R-SPN activity or behavioural performance in lean chow-fed mice, irrespective of whether they were fed or food-restricted. In contrast, in mice exposed to a high-fat diet (HFD), central TG delivery suppressed cue-evoked D2R-SPN activity in food-restricted animals without affecting reward consumption or behavioural outputs. Likewise, TG significantly reduced refeeding-evoked D2R-SPN activation in fasted HFD-mice, while leaving novelty-induced neuronal responses unchanged. Together, these findings demonstrate that obesogenic conditions reveal a metabolic state-dependent sensitivity of accumbal D2R-SPNs to circulating lipids. This work identifies TG as context-dependent modulators of D2R-SPN function and uncovers a mechanism through which dietary history interacts with current metabolic state to reshape the neural encoding of reward-predictive cues and food-directed behaviours.

The lateral habenula (LHb) integrates cortical and basal forebrain (BF) input and is implicated in negative affective states, but its involvement in negative affect associated with cannabis withdrawal is unknown. Here, effects of withdrawal from the psychoactive cannabis constituent, Δ 9 -tetrahydrocannabinol (Δ 9 -THC), on fear conditioning (FC) and LHb function were measured in rats using calcium photometry and electrophysiology. FC-associated LHb calcium signals, freezing behavior, and LHb excitatory-inhibitory (E-I) ratios were increased during Δ 9 -THC withdrawal, and local GABA release was impaired, indicating disinhibition of LHb output neurons. GABA release was also impaired at optogenetically activated BF-LHb axons, and not from ventral tegmental area-LHb inputs. The results indicate that Δ 9 -THC withdrawal increases LHb excitability through disinhibition, contributing to increased fear of primary and conditioned aversive stimuli. Our findings implicate the LHb in cannabis withdrawal-associated negative affect and offer insight into the neural mechanisms of cannabis use disorder.

Everyday decisions depend on associations between sensory stimuli, actions, and outcomes. The striatum supports these sensorimotor associations through dopamine-dependent plasticity. Recent work has characterized a local striatal microcircuit in which cholinergic interneurons (CINs) modulate dopamine release via activation of nicotinic receptors on dopamine axons. Here, we show that visual stimuli evoke dopamine in the dorsomedial striatum partly through this cholinergic mechanism. Using anatomical and functional methods to identify the pathways involved, we found that visual and auditory cortices lack connectivity with CINs and were unable to drive cholinergic-dependent dopamine release. Frontal regions, which were activated by visual stimuli, strongly recruited CINs, producing robust dopamine release both ex vivo and in vivo. These findings reveal a fundamental distinction between sensory and frontal corticostriatal inputs, demonstrating that only the latter can evoke cholinergic-dependent dopamine signals. This work establishes a framework for understanding how cortical circuits shape striatal dopamine to support reinforcement learning.

Recent work shows that the brain’s arousal system remains active during sleep, with rhythmic locus coeruleus (LC) activity shaping sleep architecture and supporting memory consolidation. The LC releases norepinephrine (NE) in infraslow (~0.02 Hz) bouts that gate NREM sleep spindles. Here, we demonstrate that heart rate (HR) fluctuations during NREM are tightly phase-locked to these NE rhythms, identifying the LC as a key driver of very-low-frequency HR variability (VLF-HRV), an understudied autonomic signal. Using optogenetics, transient LC inhibition blunts HR slowing, whereas LC activation produces rapid HR acceleration, demonstrating a direct LC-HR relationship during sleep that is maintained across a defined range of LC activity levels but breaks down when LC activity becomes excessive. We further show that infraslow HR variability is a cross-species marker of spindle-dependent memory processing. In mice, the amplitude of HR decelerations during NREM correlates with spindle activity and subsequent memory performance. Remarkably, human sleepers show the same pattern: stronger VLF-HR fluctuations during NREM correspond to increased spindle expression and better overnight memory retention. These findings reveal a mechanistic pathway through which graded changes in LC activity, up to a critical level, modulates autonomic physiology during sleep and identify infraslow HR variability as a non-invasive marker of brainstem function and memory-promoting sleep. Because LC degeneration occurs early in neurodegenerative disease, sleep-derived HR metrics may provide a scalable indicator of emerging neuromodulatory dysfunction.

Animals defend protein intake strongly, yet how protein status shapes moment-to-moment reward signalling in mesolimbic circuits remains unclear. Here, we examined how dietary protein restriction shapes nucleus accumbens (NAc) core dopamine signalling during operant feeding in the home-cage. C57BL/6 mice expressing the GRAB DA sensor in NAc core performed 1 h fixed-ratio 1 (FR1) and progressive ratio (ProgRatio) sessions to earn grain or sucrose pellets across three dietary phases; an initial non-restricted phase (NR1), a protein-restricted phase (PR), and a return to the non-restricted diet (NR2). Home-cage food intake remained stable across phases, whereas bodyweight gain was markedly reduced during PR and partially recovered in NR2. Behaviourally, operant responding increased during PR under both FR1 and ProgRatio schedules, with the most robust enhancement observed for grain pellets. Fibre photometry revealed a more selective neural effect. During FR1, pellet delivery evoked a larger NAc dopamine response for grain than for sucrose specifically during PR, whereas no pellet-type difference was detected during NR1 or NR2. By contrast, under progressive ratio, dopamine responses at pellet delivery and pellet retrieval did not differ between pellet types in any dietary phase. Together, these findings show that protein restriction does not globally amplify reward-related behaviour or dopamine signalling, but instead selectively biases NAc dopamine encoding in a manner that depends on nutritional status of the animal, nutrient context, and task demands.

Schizophrenia (SCZ) is a genetically complex neuropsychiatric disorder in which rare loss-of-function mutations in the histone methyltransferase SETD1A confer substantial risk. Although SETD1A haploinsufficiency had been linked to morphological, synaptic and behavioral abnormalities in the prefrontal cortex, whether and how SETD1A coordinates transcriptional and functional programs across different brain regions remains unknown. Here, we delineate the brain region-specific effects of SETD1A-associated dysfunction using conditional Setd1a knockout mice. We find that the dorsal striatum (dStr) and mediodorsal thalamus (MD) exhibit distinct transcriptomic and neuronal alterations to those in the PFC, and show transcriptomic enrichment for other SCZ risk genes. Loss of Setd1a in the dStr or MD drives selective vulnerability in key behavioral assays, suggesting important roles for these brain regions in the etiology of SCZ. By screening 6 existing H3K4 demethylase inhibitors, we identify the LSD1 (KDM1A) inhibitor TAK-418 as a potent modulator capable of restoring H3K4me3 levels and gene expression, as well as rescuing synaptic and SCZ-like behavioral phenotypes in the Setd1a +/− mice. Thus, our work provides a mechanistic link between high-penetrance SETD1A variants and region-specific brain dysfunction, establishing a framework for connecting rare loss-of-function variation in chromatin regulators to multidimensional neuropsychiatric phenotypes.

Social behavior varies considerably across mammalian species, yet the neural basis for these variations remains unclear. We compared the behavior of three commonly used laboratory rodent strains (C57BL/6J and CD1 mice and SD rats) in social preference and free interaction tasks. Animals carried chronically implanted electrodes in brain regions associated with social motivation. The three strains exhibited distinctly structured social behavior: SD rats showed the strongest social motivation, CD1 mice were most active, and C57BL/6J mice displayed the most restrained behavior. Electrophysiology revealed strain- and task-specific theta and gamma power and coherence. Strikingly, SD rats displayed low baseline coherence followed by robust induction during interactions, while C57BL/6J mice exhibited opposite tendencies. Machine learning models allowed accurate strain separation based solely on brain electrophysiology, with theta and gamma coherence between the prelimbic cortex and nucleus accumbens shell playing a key role in defining distinct neuronal signatures of strain-specific social strategies.

The parabrachial nucleus (PBN) is an important hub located in the pons that relays sensory signals from peripheral regions. It is genetically diverse and contains many populations that modulate feeding and responses to threatening situations. A small Ntsr1 -expressing population of neurons was identified that projects selectively to the ventromedial hypothalamus (VMH). The Ntsr1 neurons are scattered throughout the lateral PBN with a cluster of cells along the border to the nucleus of the lateral lemniscus that overlap with Cck and Foxp2 expression. Chemogenetic activation of PBN Ntsr1 neurons results in Fos induction in Nr5a1 (SF1) and Bdnf neurons in the VMH. Activation of PBN Ntsr1 neurons or their terminals in the VMH reduces food intake after fasting and increases anxiety-like behaviors. In anxiogenic feeding assays, activation of PBN Ntsr1 neurons increases latency to feed as well as reducing food intake. Photometry showed that PBN Ntsr1 -neuronal activity increases during anxiogenic situations but is suppressed during food consumption, suggesting a role in threat-induced suppression of feeding. Silencing PBN Ntsr1 neurons with tetanus toxin light-chain increased food intake and reduced anxiety. These findings reveal a genetically defined PBN to VMH circuit that responds to threats and suppresses feeding behavior.

Phasic dopamine signalling updates the value of stimuli and actions. Most empirical support for this critical process involves manipulation of extrinsic stimuli (e.g., reward magnitude, reward omission). Yet interoceptive signals clearly influence motivated behaviour. Sodium depletion generates a sodium appetite where the value of sodium increases. How the value of sodium is updated during ingestion and as animals satiate their need remains unknown. Here, we administered sodium chloride solutions via intraoral delivery and measured appetitive behaviour and dopamine release in sodium replete and deplete rats. Two different concentrations of sodium chloride were used to modulate the rate of repletion for sodium deplete rats. Appetitive behaviour was evoked by infusions only when rats were deplete, and this behavioural motif faded away during the session only when high concentration infusions were delivered. Phasic dopamine release in the nucleus accumbens had a similar pattern. We identified transition points for both behaviour and dopamine in the subgroup of rats that reached satiation and found that transitions were sharp, with dopamine decreasing just before behaviour. Moreover, transitions occurred only after enough sodium was infused to replace that typically lost upon depletion. These findings demonstrate that mesolimbic dopamine dynamically updates the value of taste stimuli based on interoceptive signals that relay satiation.

Dopamine contributes to reward-related decision making, but its contributions to decision contexts that include explicit punishment are less well understood. To elucidate the role of ventral tegmental area (VTA) dopamine neurons in decision making under risk of punishment, we used fiber photometry to record activity in these neurons during a decision-making task in which rats choose between a small, “safe” reward and a large reward associated with varying probabilities of explicit punishment. Dopamine neuron activity exhibited phasic increases during risky “Wins” (reward without punishment) and phasic decreases during risky “Losses” (reward plus punishment), each of which scaled with punishment probability and intensity. Further analyses revealed that this outcome-evoked activity predicted choices on subsequent trials. To determine whether VTA dopamine neuron activity plays a causal role in these risk-based decisions, we used optogenetics to selectively inhibit these neurons during risky Wins and Losses. Inhibition of VTA dopamine neurons during Wins selectively reduced the frequency of risky choices after a Win, whereas inhibition during Losses selectively reduced the frequency of risky choices after a Loss. These data reveal that VTA dopamine neuron activity during outcome receipt is causally related to subsequent decision-making behavior. In addition, the fact that the effects of inhibition on choice behavior were specific to the outcome with which the inhibition was paired suggests that VTA dopamine neuron activity contributes to updating after Wins and Losses independently. Significance Statement Dopamine is implicated in movement, motivation, and learning, and has been strongly linked with substance use and risk-seeking behavior. Despite these associations, it is unclear how the activity of dopamine neurons influences decision making under risk of punishment. To elucidate this, we recorded from ventral tegmental area (VTA) dopamine neurons and found that their activity at the population level integrates reward alongside the probability and intensity of punishment experienced during risky outcomes. Further, inhibition of VTA dopamine neuron activity during risky outcomes decreased risk-seeking behavior in an outcome-specific manner. Our findings align with models suggesting that VTA dopamine neurons signal prediction errors, and provide insight into the role of these neurons in adaptive decision making.

Background/Objectives: Dopamine is well known for its role in reward learning, where phasic activity encodes prediction errors and supports the formation of cue–outcome associations. More recently, dopamine has been implicated in encoding salience, novelty, and aversive events, suggesting a broader function in shaping learning and memory across motivational contexts. However, how accumbal dopamine contributes to the recovery of extinguished fear, a process central to relapse in anxiety and trauma-related disorders, remains unclear. Methods: Here, we examined NAc core dopamine dynamics during shock-induced reinstatement of an extinguished fear memory. Using dLight fiber photometry, we found that a sustained decrease in baseline dopamine marked extinction recall. In contrast, an unexpected reminder footshock prevented this reduction, maintaining dopamine levels near baseline as freezing behavior re-emerged. Results: The reminder shock also evoked a transient dopamine peak, and optogenetic manipulations demonstrated that dopamine signaling during this period bidirectionally modulated reinstatement, with enhancement of dopamine release increasing reinstatement and inhibition of dopamine terminals markedly attenuating it. Together, these results demonstrate that unexpected aversive events reset NAc core dopamine levels and gate the return of extinguished fear. Conclusions: By revealing that accumbal dopamine contributes to fear recovery, this work broadens current models of dopamine function and identifies a neural mechanism through which surprising events may promote relapse of aversive memories in anxiety- and stress-related disorders.

Current disease-sensing devices primarily focus on distinguishing between healthy and diseased states, effective for diagnosis but limited in guiding optimal intervention timing for prevention. We developed a tripartite framework identifying pre-disease state in depression, a reversible phase preceding irreversible onset. Using complex systems theory, we analyzed early-warning signals emerging as biological systems approach critical transitions. Continuous monitoring of nine multimodal biomarkers—spanning electrophysiological, behavioral, and biological—enabled classification into normal, pre-disease, and disease states by quantitatively defining critical points. An artificial intelligence agent classified disease states with 95.2% accuracy using multimodal data, enabled by ultrasoft neural probes for stable, low-damage recordings. Therapeutic validation with a skin-attachable wireless vagus nerve stimulator integrating soft three-dimensional electrodes demonstrated superior efficacy during pre-disease states. Subjects treated during pre-disease showed faster recovery and greater therapeutic responses, while those treated after disease onset failed to achieve full recovery. This framework provides evidence-based rationale for early intervention.

Addicted individuals prefer drug rewards over natural rewards. Dysregulated dopamine (DA) signaling in the nucleus accumbens is thought to trigger this phenomenon, but there is little supporting evidence. Using a genetically encoded sensor, we monitored DA dynamics in mice in an operant associative task with natural (fat solution) or artificial (cocaine or optogenetic DA self-stimulation) rewards. During learning, DA transients emerged at predictive cues for both reward types, vanished at natural reward delivery, but persisted at artificial reward delivery. In choice sessions, animals’ preferences varied from exclusive natural reward to exclusive artificial reward. Individual preferences were predicted by cue-evoked DA response magnitudes. Revaluation of reward contingencies shifted reward-type preference and DA signaling accordingly, but not in mice showing compulsive drug seeking, indicating impaired value updating in these mice. These results support a model in which DA drives adaptive reward seeking, with persistent cue-associated DA signaling underpinning addiction vulnerability.

Exposure to stress can cause long-lasting enhancement of fear and other defensive responses that extend beyond the cues or contexts associated with the original traumatic event. These nonassociative consequences of stress, referred to as fear sensitization, are thought to underlie some symptoms of trauma-related disorders. Fear sensitization has been predominantly studied using the stress-enhanced fear learning (SEFL) paradigm, which models the stress-induced amplification of fear learning. Less is known about the mechanisms through which unlearned fear responses are sensitized by stress. Here, we investigated the neural mechanisms for sensitization of unlearned fear responses using a paradigm we termed stress-enhanced fear responding (SEFR). In this model, mice exposed to a single session of footshock stress exhibit enhanced freezing to a novel tone stimulus. To investigate brain regions that might mediate SEFR, we first used c-Fos mapping to identify neural activity changes associated with stress-induced enhancement of unlearned fear. Our c-Fos screen identified the posterior paraventricular thalamus (pPVT) as a region that was persistently hyperactive after footshock stress and whose activity correlated with behavioral expression of SEFR. Using fiber photometry, we observed that SEFR, but not SEFL, was associated with increased activity in the pPVT. Next, we found that chemogenetic inhibition of the pPVT blocked both the induction of SEFR during stress and its later expression, while artificial stimulation of pPVT in stress-naive mice was sufficient to recapitulate SEFR. Interestingly, pPVT inhibition or stimulation did not affect acquisition or expression of SEFL. In conclusion, our results indicate that sensitization of fear learning (SEFL) and sensitization of unlearned fear (SEFR) have distinct neural mechanisms. Our results identify pPVT hyperactivity as a mechanism for stress-induced sensitization of unlearned fear and highlight pPVT as a potential target for treating arousal and reactivity symptoms of trauma- and stressor-related disorders.

The basolateral amygdala (BLA) is important for assigning emotional weight to environmental stimuli to guide learning, discrimination, and adaptive vigilance. The contribution of neuromodulator inputs such as dopamine to this process is unclear. Here, we recorded in vivo BLA dopamine signals across emotional learning paradigms. We show that BLA dopamine scales with a stimulus's emotional intensity, but not value or associative strength, and reward-predictive cues evoke BLA dopamine signals that diminish, rather than grow, with training. In a complex learning context, where rats discriminated between multiple valenced associations, the magnitude of cue-evoked BLA dopamine re-scaled to report the level of relative emotional saliency. Threat and safety cues prompted larger, sustained dopamine signals compared to reward and neutral cues. Together, our data reveal a pattern of dopamine encoding with a unique contribution to learning, for dynamic disambiguation of affective states.

Adaptive behavior requires that behaviorally relevant signals gain access to neural circuits guiding action. The thalamus has long been proposed to regulate information flow to cortical and subcortical systems, yet whether it also tracks internally generated goal signals remains unclear. Here, we show that neurons in the paraventricular thalamus (PVT) projecting to the nucleus accumbens (NAc) encode the motivational value of safety during active avoidance. As mice learn to avoid threat, PVT→NAc neurons develop a signal that emerges selectively at successful avoidance, is experience-dependent, and diminishes following outcome devaluation. Selective silencing of the PVT→NAc pathway at safety onset reduces the motivational value assigned to safety without impairing action–outcome learning. Mechanistically, PVT input engages cholinergic interneurons (CIN) in the NAc to regulate dopamine release via synaptic potentiation mediated by GluA2-lacking AMPA receptors at PVT–CIN synapses. Disrupting this plasticity reduces the motivational impact of safety. These findings identify a thalamostriatal mechanism through which learned goals gain stable access to motivational circuitry.

Social recognition, the ability to distinguish between individuals, is essential for cognitively demanding social behaviors. The anterior olfactory nucleus (AON), a primary olfactory cortical region, is implicated in this process, but the underlying neurocircuitry remains poorly understood. Here, we generated a novel mouse line to enable genetic access to AON pyramidal neurons and mapped their whole-brain synaptic inputs and outputs. The medial prefrontal cortex (mPFC), a crucial hub for social cognition, is the primary neocortical target of AON neurons, which form monosynaptic excitatory connections with a substantial fraction of mPFC neurons. The AONmPFC pathway is activated during social investigation, and chemogenetic inhibition of this pathway impairs social recognition. Moreover, an analogous AON-prefrontal pathway is present in humans, as supported by resting-state functional magnetic resonance imaging (fMRI) functional connectivity analyses. Taken together, these findings reveal a conserved olfactory-prefrontal circuit spanning mice to humans, potentially linking olfactory dysfunction to neuropsychiatric disorders.

The medial nucleus accumbens shell (medNAcSh) is a key regulator of hedonic feeding, controlling reward consumption through its projections to downstream structures. Recent studies showed that the primary cellular mediators of these effects are dopamine 1 receptor-positive striatal projection neurons (D1-SPNs). Specifically, D1-SPN activity gets inhibited during reward consumption, and such inhibition is necessary and sufficient to authorize consumption, independent of metabolic need. Anatomically, the medNAcSh spans 1–1.5 mm along the rostro-caudal axis in mice, and previous studies have reported functional gradients along this axis. For instance, pharmacological studies have suggested that rostral rather than caudal medNAcSh regulates appetitive behavior. However, the mechanisms underlying this topographical gradient remain unknown. Here, we hypothesized that D1-SPNs contribute to this gradient by regulating hedonic feeding via a specific hotspot in the rostral medNAcSh. Using calcium monitoring with fiber photometry in mice, we show that rostral medNAcSh D1-SPNs demonstrate inhibitory responses during reward consumption, while caudal D1-SPNs do not. Consistently, optogenetic stimulation of rostral D1-SPNs inhibits consumption, while stimulation of caudal D1-SPNs had minimal effects, confirming the existence of a functional rostro-caudal gradient. Importantly, we observed no differences between rostral and caudal D1-SPNs in their responses to aversive stimuli, suggesting that the D1-SPN gradient is specific to appetitive contexts. To investigate potential molecular correlates of this functional gradient, we leveraged open-source anatomy datasets and performed fluorescent in situ hybridization, identifying Stard5 and Peg10 as markers enriched in the rostral and caudal medNAcSh, respectively. Finally, we developed a novel Stard5-Flp driver line to selectively target the rostral hotspot and demonstrated that Stard5+ neurons recapitulate rostral D1-SPN activity patterns. Together, these findings establish a spatially confined rostral medNAcSh subregion as a critical regulator of reward consumption and introduce Stard5 as a molecular tool for its manipulation, offering new opportunities for intervention in dysregulated eating.

Dopamine is essential for striatal function and learning. Striatal dopamine release can be triggered by dopamine cell firing, but also by coordinated cholinergic interneuron activity, which stimulates dopamine release via presynaptic nicotinic acetylcholine receptors on dopamine axons. While acetylcholine-dependent dopamine release is well-documented ex vivo and under artificial optogenetic stimulation in vivo , its role during natural behavior has remained unclear. One possible endogenous driver of acetylcholine-dependent dopamine release is thalamic input, which provides strong excitatory drive to cholinergic interneurons. To examine whether thalamic input provokes acetylcholine-dependent dopamine release during behavior, we performed simultaneous fiber photometry recordings of striatal dopamine (GRAB-rDA3m) and thalamic axon activity (gCaMP8m) in the dorsomedial (DMS) and dorsolateral striatum (DLS) of mice learning the accelerating rotarod, a striatal-dependent task that demands precise and effortful motor control. Recordings were obtained on- and off-task and across days of training to capture the full arc of learning. Dopamine transients in DMS, but not DLS, were frequently coupled to peaks in thalamic axon activity via an acetylcholine-dependent mechanism. The occurrence of these thalamic-evoked DMS dopamine transients depended on learning, task engagement, and the recent history of dopamine activity, but did not contribute to motor error signals. Together, these findings establish thalamic input as a physiological driver of acetylcholine-dependent dopamine release in DMS. Moreover, they reveal that striatal sensitivity to this local release mechanism is dynamically gated by dopaminergic history, providing a compelling framework for understanding how local and soma-triggered dopamine signals are coordinated to support learning.

Quickly adapting behavior to changing environments is key for survival. Brain dopamine signals reinforce actions that procure rewards and avoid adverse outcomes. However, the molecular mechanisms within the circuits that actuate these signals at subsecond timescales remain unexplored. We show that midbrain endocannabinoid mobilization is the conditio sine qua non for behavioral invigoration evoked by external cues. Reward- and punishment-predictive signals require intact retrograde release of the endocannabinoid 2-arachidonoylglycerol (2-AG) from ventral tegmental area dopamine neurons. Moreover, we show that genetic deletion of presynaptic cannabinoid type-1 receptors (CB1Rs), the target for 2-AG, recapitulates behavioral and dopaminergic deficits. Exploiting causal inference tools to interpret continuous brain signals, we describe that 2-AG/CB1R communication is essential for striatal dopamine release to signal salient cues and propel conditioned responding. These findings reveal an endocannabinoid-mediated disinhibition mechanism exploited by dopamine neurons to orchestrate striatal dopamine release events, enabling adaptive action selection informed by cues.