Dynamical prefrontal population coding during defensive behaviours

Distinct prelimbic cortex neuronal responses to emotional states of others drive emotion recognition in adult mice

The ability to perceive the emotional states of others, termed emotion recognition, allows individuals to adapt their conduct to the social environment. The brain mechanisms underlying this capacity, known to be impaired in individuals with autism spectrum disorder (ASD), remain, however, elusive. Here, we show that adult mice can discern between emotional states of conspecifics. Fiber photometry recordings of calcium signals in the prelimbic (PrL) medial prefrontal cortex revealed inhibition of pyramidal neurons during investigation of emotionally aroused individuals, as opposed to transient excitation toward naive conspecifics. Chronic electrophysiological recordings at the single-cell level indicated social stimulus-specific responses in PrL neurons at the onset and conclusion of social investigation bouts, potentially regulating the initiation and termination of social interactions. Finally, optogenetic augmentation of the differential neuronal response enhanced emotion recognition, while its reduction eliminated such behavior. Thus, differential PrL neuronal response to individuals with distinct emotional states underlies murine emotion recognition.

Persistent Threat Avoidance Following Negative Reinforcement Is Not Associated with Elevated State Anxiety

Obsessive-compulsive disorder (OCD) is a debilitating illness consisting of obsessions and compulsions. OCD severity and treatment response are correlated with avoidant behaviors thought be performed to alleviate obsession-related anxiety. However, little is known about either the role of avoidance in the development of OCD or the interplay between anxiety states and avoidance behaviors. We have developed an instrumental negative reinforcement (i.e., active avoidance) paradigm in which mice must lever press to avoid upcoming footshocks. We show that mice (both sexes) can learn this task with high acquisition rates (75%) and that this behavior is largely stable when introducing uncertainty and modifying task structure. Furthermore, mice continue to perform avoidance responses on trials where lever pressing is not reinforced and increase response rates as they are maintained on this paradigm. With this paradigm, we did not find a relationship between negative reinforcement history and anxiety-related behaviors in well-established anxiety assays. Finally, we performed exploratory analyses to identify candidate regions involved in well-trained negative reinforcement using expression of the immediate early gene c-Fos. We detected correlated c-Fos expression in (1) corticostriatal regions which regulate active avoidance in other paradigms and (2) amygdala circuits known to regulate conditioned defensive behaviors.

Prefrontal dopamine activity is critical for rapid threat avoidance learning

The medial prefrontal cortex (mPFC) is required for learning associations that determine whether animals approach or avoid potential threats in the environment. Dopaminergic (DA) projections from the ventral tegmental area (VTA) to the mPFC carry information, particularly about aversive outcomes, that may inform prefrontal computations. But the role of prefrontal DA in learning based on aversive outcomes remains poorly understood. Here, we used platform mediated avoidance (PMA) to study the role of mPFC DA in threat avoidance learning in mice. We show that activity in VTA-mPFC dopaminergic terminals is required for avoidance learning, but not for escape, conditioned fear, or to recall a previously learned avoidance strategy. mPFC DA is most dynamic in the early stages of learning, and encodes aversive outcomes, their omissions, and threat-induced behaviors. Computational models of PMA behavior and DA activity revealed that mPFC DA influences learning rates and encodes the predictive relationships between cues and adaptive behaviors. Taken together, these data indicate that mPFC DA is necessary to rapidly learn behaviors required to avoid signaled threats, but not for learning cue-threat associations.

ErbB4 precludes the occurrence of PTSD-like fear responses by supporting the bimodal activity of the central amygdala

Post-traumatic stress disorder (PTSD) often arises after exposure to traumatic events and is characterized by dysregulated fear responses. Although the associations of erb-b2 receptor tyrosine kinase 4 (ErbB4) with various neuropsychiatric diseases, including schizophrenia and bipolar disorder, have been widely examined, the physiological roles of ErbB4 in PTSD and fear responses remain unclear. Using Cre-dependent ErbB4 knockout (KO) mice, we observed that PTSD-like fear behaviors emerged in ErbB4-deficient mice, particularly in inhibitory neurons. Specifically, the loss of ErbB4 in somatostatin-expressing (SST+) neurons was sufficient to induce PTSD-like fear responses. We also adopted the CRISPR/Cas9 system for region-specific KO of ErbB4, which revealed that ErbB4 deletion in SST+ neurons of the lateral division of the amygdala (CeL) caused elevated anxiety and PTSD-like fear generalization. Consistent with its physiological role, ErbB4 expression was diminished in CeLSST neurons from mice that exhibited PTSD-like phenotypes. While fear On and Off cells identified in the CeL displayed distinct responses to conditioned and novel cues, as previously shown, the selectivity of those On and Off cells was compromised in SSTErbB4-/- and stressed mice, which displayed strong fear generalization. Therefore, the bimodal activity that CeL On/Off cells display is likely required for proper discrimination of fearful stimuli from ambient stimuli, which should be sustained by the presence of ErbB4. Taken together, our data substantiate the correlation between PTSD-like fear responses and ErbB4 expression in CeLSST neurons and further underscore the functional effects of ErbB4 in CeLSST neurons, supporting the bimodal responses of CeL neurons.

Neural decoding and feature selection methods for closed-loop control of avoidance behavior

Objective.Many psychiatric disorders involve excessive avoidant or defensive behavior, such as avoidance in anxiety and trauma disorders or defensive rituals in obsessive-compulsive disorders. Developing algorithms to predict these behaviors from local field potentials (LFPs) could serve as the foundational technology for closed-loop control of such disorders. A significant challenge is identifying the LFP features that encode these defensive behaviors.Approach.We analyzed LFP signals from the infralimbic cortex and basolateral amygdala of rats undergoing tone-shock conditioning and extinction, standard for investigating defensive behaviors. We utilized a comprehensive set of neuro-markers across spectral, temporal, and connectivity domains, employing SHapley Additive exPlanations for feature importance evaluation within Light Gradient-Boosting Machine models. Our goal was to decode three commonly studied avoidance/defensive behaviors: freezing, bar-press suppression, and motion (accelerometry), examining the impact of different features on decoding performance.Main results.Band power and band power ratio between channels emerged as optimal features across sessions. High-gamma (80-150 Hz) power, power ratios, and inter-regional correlations were more informative than other bands that are more classically linked to defensive behaviors. Focusing on highly informative features enhanced performance. Across 4 recording sessions with 16 subjects, we achieved an average coefficient of determination of 0.5357 and 0.3476, and Pearson correlation coefficients of 0.7579 and 0.6092 for accelerometry jerk and bar press rate, respectively. Utilizing only the most informative features revealed differential encoding between accelerometry and bar press rate, with the former primarily through local spectral power and the latter via inter-regional connectivity. Our methodology demonstrated remarkably low training/inference time and memory usage, requiring<310 ms for training,<0.051 ms for inference, and 16.6 kB of memory, using a single core of AMD Ryzen Threadripper PRO 5995WX CPU.Significance.Our results demonstrate the feasibility of accurately decoding defensive behaviors with minimal latency, using LFP features from neural circuits strongly linked to these behaviors. This methodology holds promise for real-time decoding to identify physiological targets in closed-loop psychiatric neuromodulation.

Locus coeruleus modulation of prefrontal dynamics and encoding of flexible rule switching

Behavioral flexibility, the ability to adjust behavioral strategies in response to changing environmental contingencies and internal demands, is fundamental to cognitive functions. Despite a large body of pharmacology and lesion studies, the underlying neurophysiological correlates and mechanisms that support flexible rule switching are under active investigation. To address this question, we trained mice to distinguish complex sensory cues comprising different perceptual dimensions (set shifting). Endoscopic calcium imaging revealed that medial prefrontal cortex (mPFC) neurons exhibited pronounced dynamic changes during rule switching. Notably, prominent encoding capacity in the mPFC was associated with switching across, but not within perceptual dimensions. We then showed the functional importance of the ascending input from the locus coeruleus (LC), as LC inhibition impaired rule switching behavior and impeded mPFC dynamic processes and encoding. Our results highlight the pivotal role of the mPFC in set shifting processes and demonstrate the profound impact of ascending neuromodulation on shaping prefrontal neural dynamics and behavioral flexibility.

Circadian Rhythms in Conditioned Threat Extinction Reflect Time-of-Day Differences in Ventromedial Prefrontal Cortex Neural Processing

Circadian rhythms in conditioned threat extinction emerge from a tissue-level circadian timekeeper, or local clock, in the ventromedial prefrontal cortex (vmPFC). Yet it remains unclear how this local clock contributes to extinction-dependent adaptations. Here we used single-unit and local field potential analyses to interrogate neural activity in the male rat vmPFC during repeated extinction sessions at different times of day. In association with superior recall of a remote extinction memory during the circadian active phase, vmPFC putative principal neurons exhibited phasic firing that was amplified for cue presentations and diminished at transitions in freezing behavior. Coupling of vmPFC gamma amplitude to the phase of low-frequency oscillations was greater during freezing than mobility, and this difference was augmented during the active phase, highlighting a time-of-day dependence in the organization of freezing- versus mobility-associated cell assemblies. Additionally, a greater proportion of vmPFC neurons were phase-locked to low-frequency oscillations during the active phase, consistent with heightened neural excitability at this time of day. Our results suggest that daily fluctuations in vmPFC excitability precipitate enhanced neural recruitment into extinction-based cell assemblies during the active phase, providing a potential mechanism by which the vmPFC local clock modulates circuit and behavioral plasticity during conditioned threat extinction.

Population-level coding of avoidance learning in medial prefrontal cortex

The medial prefrontal cortex (mPFC) has been proposed to link sensory inputs and behavioral outputs to mediate the execution of learned behaviors. However, how such a link is implemented has remained unclear. To measure prefrontal neural correlates of sensory stimuli and learned behaviors, we performed population calcium imaging during a new tone-signaled active avoidance paradigm in mice. We developed an analysis approach based on dimensionality reduction and decoding that allowed us to identify interpretable task-related population activity patterns. While a large fraction of tone-evoked activity was not informative about behavior execution, we identified an activity pattern that was predictive of tone-induced avoidance actions and did not occur for spontaneous actions with similar motion kinematics. Moreover, this avoidance-specific activity differed between distinct avoidance actions learned in two consecutive tasks. Overall, our results are consistent with a model in which mPFC contributes to the selection of goal-directed actions by transforming sensory inputs into specific behavioral outputs through distributed population-level computations.

Behavioral outputs and overlapping circuits between conditional fear and active avoidance

Aversive learning can produce a wide variety of defensive behavioral responses depending on the circumstances, ranging from reactive responses like freezing to proactive avoidance responses. While most of this initial learning is behaviorally supported by an expectancy of an aversive outcome and neurally supported by activity within the basolateral amygdala, activity in other brain regions become necessary for the execution of defensive strategies that emerge in other aversive learning paradigms such as active avoidance. Here, we review the neural circuits that support both reactive and proactive defensive behaviors that are motivated by aversive learning, and identify commonalities between the neural substrates of these distinct (and often exclusive) behavioral strategies.

Transformations in prefrontal ensemble activity underlying rapid threat avoidance learning

The capacity to learn cues that predict aversive outcomes, and understand how to avoid those outcomes, is critical for adaptive behavior. Naturalistic avoidance often means accessing a safe location, but whether a location is safe depends on the nature of the impending threat. These relationships must be rapidly learned if animals are to survive. The prelimbic subregion (PL) of the medial prefrontal cortex (mPFC) integrates learned associations to influence these threat avoidance strategies. Prior work has focused on the role of PL activity in avoidance behaviors that are fully established, leaving the prefrontal mechanisms that drive rapid avoidance learning poorly understood. To determine when and how these learning-related changes emerge, we recorded PL neural activity using miniscope calcium imaging as mice rapidly learned to avoid a threatening cue by accessing a safe location. Over the course of learning, we observed enhanced modulation of PL activity representing intersections of a threatening cue with safe or risky locations and movements between them. We observed rapid changes in PL population dynamics that preceded changes observable in the encoding of individual neurons. Successful avoidance could be predicted from cue-related population dynamics during early learning. Population dynamics during specific epochs of the conditioned tone period correlated with the modeled learning rates of individual animals. In contrast, changes in single-neuron encoding occurred later, once an avoidance strategy had stabilized. Together, our findings reveal the sequence of PL changes that characterize rapid threat avoidance learning.

Distinct roles of prefrontal cortex neurons in set shifting

Cognitive flexibility, the ability to adjust behavioral strategies in response to changing environmental contingencies, requires adaptive processing of internal states and contextual cues to guide goal-oriented behavior, and is dependent on prefrontal cortex (PFC) functions. However, the neurophysiological underpinning of how the PFC supports cognitive flexibility is not well understood and has been under active investigation. We recorded spiking activity from single PFC neurons in mice performing the attentional set-shifting task, where mice learned to associate different contextually relevant sensory stimuli to reward. We identified subgroups of PFC neurons encoding task context, choice and trial outcome. Putative fast-spiking neurons were more involved in representing outcome and choice than putative regular-spiking neurons. Regression model further revealed that task context and trial outcome modulated the activity of choice-encoding neurons in rule-dependent and cell type-dependent manners. Together, our data provide new evidence to elucidate PFC's role in cognitive flexibility, suggesting differential cell type-specific engagement during set shifting, and that both contextual rule representation and trial outcome monitoring underlie PFC's unique capacity to support flexible behavioral switching.

Convergent direct and indirect cortical streams shape avoidance decisions in mice via the midline thalamus

Current concepts of corticothalamic organization in the mammalian brain are mainly based on sensory systems, with less focus on circuits for higher-order cognitive functions. In sensory systems, first-order thalamic relays are driven by subcortical inputs and modulated by cortical feedback, while higher-order relays receive strong excitatory cortical inputs. The applicability of these principles beyond sensory systems is uncertain. We investigated mouse prefronto-thalamic projections to the midline thalamus, revealing distinct top-down control. Unlike sensory systems, this pathway relies on indirect modulation via the thalamic reticular nucleus (TRN). Specifically, the prelimbic area, which influences emotional and motivated behaviors, impacts instrumental avoidance responses through direct and indirect projections to the paraventricular thalamus. Both pathways promote defensive states, but the indirect pathway via the TRN is essential for organizing avoidance decisions through disinhibition. Our findings highlight intra-thalamic circuit dynamics that integrate cortical cognitive signals and their role in shaping complex behaviors.

Threat-dependent scaling of prelimbic dynamics to enhance fear representation

Promptly identifying threatening stimuli is crucial for survival. Freezing is a natural behavior displayed by rodents toward potential or actual threats. Although it is known that the prelimbic cortex (PL) is involved in both risk evaluation and in fear and anxiety-like behavior expression, here we explored whether PL neuronal activity can dynamically represent different internal states of the same behavioral output (i.e., freezing). We found that freezing can always be decoded from PL activity at a population level. However, the sudden presentation of a fearful stimulus quickly reshaped the PL to a new neuronal activity state, an effect not observed in other cortical or subcortical regions examined. This shift changed PL freezing representation and is necessary for fear memory expression. Our data reveal the unique role of the PL in detecting threats and internally adjusting to distinguish between different freezing-related states in both unconditioned and conditioned fear representations.

A prefrontal motor circuit initiates persistent movement

Persistence reinforces continuous action, which benefits animals in many aspects. Diverse external or internal signals may trigger animals to start a persistent movement. However, it is unclear how the brain decides to persist with current actions by selecting specific information. Using single-unit extracellular recordings and opto-tagging in awake mice, we demonstrated that a group of dorsal mPFC (dmPFC) motor cortex projecting (MP) neurons initiate a persistent movement by selectively encoding contextual information rather than natural valence. Inactivation of dmPFC MP neurons impairs the initiation and reduces neuronal activity in the insular and motor cortex. After the persistent movement is initiated, the dmPFC MP neurons are not required to maintain it. Finally, a computational model suggests that a successive sensory stimulus acts as an input signal for the dmPFC MP neurons to initiate a persistent movement. These results reveal a neural initiation mechanism on the persistent movement.

The prelimbic prefrontal cortex mediates the development of lasting social phobia as a consequence of social threat conditioning

Social phobia is highly detrimental for social behavior, mental health, and productivity. Despite much previous research, the behavioral and neurobiological mechanisms associated with the development of social phobia remain elusive. To investigate these issues, the present study implemented a mouse model of social threat conditioning in which mice received electric shock punishment upon interactions with unfamiliar conspecifics. This resulted in immediate reductions in social behavior and robust increases in defensive mechanisms such as avoidance, freezing, darting, and ambivalent stretched posture. Furthermore, social deficits lasted for prolonged periods and were independent of contextual settings, sex variables, or particular identity of the social stimuli. Shedding new light into the neurobiological factors contributing to this phenomenon, we found that optogenetic silencing of the prelimbic (PL), but not the infralimbic (IL), subregion of the medial prefrontal cortex (mPFC) during training led to subsequent forgetting and development of lasting social phobia. Similarly, pharmacological inhibition of NMDARs in PL also impaired the development of social phobia. These findings are consistent with the notion that social-related trauma is a prominent risk factor for the development of social phobia, and that this phenomenon engages learning-related mechanisms within the prelimbic prefrontal cortex to promote prolonged representations of social threat.

Abstract Figure

Measuring the replicability of our own research

In the study of transgenic mouse models of neurodevelopmental and neurodegenerative disorders, we use batteries of tests to measure deficits in behaviour and from the results of these tests, we make inferences about the mental states of the mice that we interpret as deficits in "learning", "memory", "anxiety", "depression", etc. This paper discusses the problems of determining whether a particular transgenic mouse is a valid mouse model of disease X, the problem of background strains, and the question of whether our behavioural tests are measuring what we say they are. The problem of the reliability of results is then discussed: are they replicable between labs and can we replicate our results in our own lab? This involves the study of intra- and inter- experimenter reliability. The variables that influence replicability and the importance of conducting a complete behavioural phenotype: sensory, motor, cognitive and social emotional behaviour are discussed. Then the thorny question of failure to replicate is examined: Is it a curse or a blessing? Finally, the role of failure in research and what it tells us about our research paradigms is examined.

The dynamic state of a prefrontal-hypothalamic-midbrain circuit commands behavioral transitions

Innate behaviors meet multiple needs adaptively and in a serial order, suggesting the existence of a hitherto elusive brain dynamics that brings together representations of upcoming behaviors during their selection. Here we show that during behavioral transitions, possible upcoming behaviors are encoded by specific signatures of neuronal populations in the lateral hypothalamus (LH) that are active near beta oscillation peaks. Optogenetic recruitment of intrahypothalamic inhibition at this phase eliminates behavioral transitions. We show that transitions are elicited by beta-rhythmic inputs from the prefrontal cortex that spontaneously synchronize with LH 'transition cells' encoding multiple behaviors. Downstream of the LH, dopamine neurons increase firing during beta oscillations and also encode behavioral transitions. Thus, a hypothalamic transition state signals alternative future behaviors, encodes the one most likely to be selected and enables rapid coordination with cognitive and reward-processing circuitries, commanding adaptive social contact and eating behaviors.

The representational geometry of emotional states in basolateral amygdala

Sensory stimuli associated with aversive outcomes cause multiple behavioral responses related to an animal's evolving emotional state, but neural mechanisms underlying these processes remain unclear. Here aversive stimuli were presented to mice, eliciting two responses reflecting fear and flight to safety: tremble and ingress into a virtual burrow. Inactivation of basolateral amygdala (BLA) eliminated differential responses to aversive and neutral stimuli without eliminating responses themselves, suggesting BLA signals valence, not motor commands. However, two-photon imaging revealed that neurons typically exhibited mixed selectivity for stimulus identity, valence, tremble and/or ingress. Despite heterogeneous selectivity, BLA representational geometry was lower-dimensional when encoding valence, tremble and safety, enabling generalization of emotions across conditions. Further, tremble and valence coding directions were orthogonal, allowing linear readouts to specialize. Thus BLA representational geometry confers two computational properties that identify specialized neural circuits encoding variables describing emotional states: generalization across conditions, and readouts lacking interference from other readouts.

A prefrontal motor circuit initiates persistent movement

Persistence reinforces continuous action, which benefits animals in many aspects. Diverse information may trigger animals to start a persistent movement. However, it is unclear how the brain decides to persist with current actions by selecting specific information. Using single-unit extracellular recordings and opto-tagging in awake mice, we demonstrated that a group of dorsal mPFC (dmPFC) motor cortex projecting (MP) neurons initiate a persistent movement selectively encoding contextual information rather than natural valence. Inactivation of dmPFC MP neurons impairs the initiation and reduces neuronal activity in the insular and motor cortex. Finally, a computational model suggests that a successive sensory stimulus acts as an input signal for the dmPFC MP neurons to initiate a persistent movement. These results reveal a neural initiation mechanism on the persistent movement.

Prefrontal Regulation of Safety Learning during Ethologically Relevant Thermal Threat

Learning and adaptation during sources of threat and safety are critical mechanisms for survival. The prelimbic (PL) and infralimbic (IL) subregions of the medial prefrontal cortex (mPFC) have been broadly implicated in the processing of threat and safety. However, how these regions regulate threat and safety during naturalistic conditions involving thermal challenge still remains elusive. To examine this issue, we developed a novel paradigm in which adult mice learned that a particular zone that was identified with visuospatial cues was associated with either a noxious cold temperature ("threat zone") or a pleasant warm temperature ("safety zone"). This led to the rapid development of avoidance behavior when the zone was paired with cold threat or approach behavior when the zone was paired with warm safety. During a long-term test without further thermal reinforcement, mice continued to exhibit robust avoidance or approach to the zone of interest, indicating that enduring spatial-based memories were formed to represent the thermal threat and thermal safety zones. Optogenetic experiments revealed that neural activity in PL and IL was not essential for establishing the memory for the threat zone. However, PL and IL activity bidirectionally regulated memory formation for the safety zone. While IL activity promoted safety memory during normal conditions, PL activity suppressed safety memory especially after a stress pretreatment. Therefore, a working model is proposed in which balanced activity between PL and IL is favorable for safety memory formation, whereas unbalanced activity between these brain regions is detrimental for safety memory after stress.

A distinct cortical code for socially learned threat

Animals can learn about sources of danger while minimizing their own risk by observing how others respond to threats. However, the distinct neural mechanisms by which threats are learned through social observation (known as observational fear learning1-4 (OFL)) to generate behavioural responses specific to such threats remain poorly understood. The dorsomedial prefrontal cortex (dmPFC) performs several key functions that may underlie OFL, including processing of social information and disambiguation of threat cues5-11. Here we show that dmPFC is recruited and required for OFL in mice. Using cellular-resolution microendoscopic calcium imaging, we demonstrate that dmPFC neurons code for observational fear and do so in a manner that is distinct from direct experience. We find that dmPFC neuronal activity predicts upcoming switches between freezing and moving state elicited by threat. By combining neuronal circuit mapping, calcium imaging, electrophysiological recordings and optogenetics, we show that dmPFC projections to the midbrain periaqueductal grey (PAG) constrain observer freezing, and that amygdalar and hippocampal inputs to dmPFC opposingly modulate observer freezing. Together our findings reveal that dmPFC neurons compute a distinct code for observational fear and coordinate long-range neural circuits to select behavioural responses.

PREFRONTAL CORRELATES OF FEAR GENERALIZATION DURING ENDOCANNABINOID DEPLETION

Maladaptive fear generalization is one of the hallmarks of trauma-related disorders. The endocannabinoid 2-arachidonoylglycerol (2-AG) is crucial for modulating anxiety, fear, and stress adaptation but its role in balancing fear discrimination versus generalization is not known. To address this, we used a combination of plasma endocannabinoid measurement and neuroimaging from a childhood maltreatment exposed and non-exposed mixed population combined with human and rodent fear conditioning models. Here we show that 2-AG levels are inversely associated with fear generalization at the behavioral level in both mice and humans. In mice, 2-AG depletion increases the proportion of neurons, and the similarity between neuronal representations, of threat-predictive and neutral stimuli within prelimbic prefrontal cortex ensembles. In humans, increased dorsolateral prefrontal cortical-amygdala resting state connectivity is inversely correlated with fear generalization. These data provide convergent cross-species evidence that 2-AG is a key regulator of fear generalization and suggest 2-AG deficiency could represent a trauma-related disorder susceptibility endophenotype.

Neuronal tuning to threat exposure remains stable in the mouse prefrontal cortex over multiple days

Intense threat elicits action in the form of active and passive coping. The medial prefrontal cortex (mPFC) executes top-level control over the selection of threat coping strategies, but the dynamics of mPFC activity upon continuing threat encounters remain unexplored. Here, we used 1-photon calcium imaging in mice to probe the activity of prefrontal pyramidal cells during repeated exposure to intense threat in a tail suspension (TS) paradigm. A subset of prefrontal neurons displayed selective activation during TS, which was stably maintained over days. During threat, neurons showed specific tuning to active or passive coping. These responses were unrelated to general motion tuning and persisted over days. Moreover, the neural manifold traversed by low-dimensional population activity remained stable over subsequent days of TS exposure and was preserved across individuals. These data thus reveal a specific, temporally, and interindividually conserved repertoire of prefrontal tuning to behavioral responses under threat.

Stable coding of aversive associations in medial prefrontal populations

The medial prefrontal cortex (mPFC) is at the core of numerous psychiatric conditions, including fear and anxiety-related disorders. Whereas an abundance of evidence suggests a crucial role of the mPFC in regulating fear behaviour, the precise role of the mPFC in this process is not yet entirely clear. While studies at the single-cell level have demonstrated the involvement of this area in various aspects of fear processing, such as the encoding of threat-related cues and fear expression, an increasingly prevalent idea in the systems neuroscience field is that populations of neurons are, in fact, the essential unit of computation in many integrative brain regions such as prefrontal areas. What mPFC neuronal populations represent when we face threats? To address this question, we performed electrophysiological single-unit population recordings in the dorsal mPFC while mice faced threat-predicting cues eliciting defensive behaviours, and performed pharmacological and optogenetic inactivations of this area and the amygdala. Our data indicated that the presence of threat-predicting cues induces a stable coding dynamics of internally driven representations in the dorsal mPFC, necessary to drive learned defensive behaviours. Moreover, these neural population representations primary reflect learned associations rather than specific defensive behaviours, and the construct of such representations relies on the functional integrity of the amygdala.

Prefrontal circuits encode both general danger and specific threat representations

Behavioral adaptation to potential threats requires both a global representation of danger to prepare the organism to react in a timely manner but also the identification of specific threatening situations to select the appropriate behavioral responses. The prefrontal cortex is known to control threat-related behaviors, yet it is unknown whether it encodes global defensive states and/or the identity of specific threatening encounters. Using a new behavioral paradigm that exposes mice to different threatening situations, we show that the dorsomedial prefrontal cortex (dmPFC) encodes a general representation of danger while simultaneously encoding a specific neuronal representation of each threat. Importantly, the global representation of danger persisted in error trials that instead lacked specific threat identity representations. Consistently, optogenetic prefrontal inhibition impaired overall behavioral performance and discrimination of different threatening situations without any bias toward active or passive behaviors. Together, these data indicate that the prefrontal cortex encodes both a global representation of danger and specific representations of threat identity to control the selection of defensive behaviors.

Early selection of task-relevant features through population gating

Brains can gracefully weed out irrelevant stimuli to guide behavior. This feat is believed to rely on a progressive selection of task-relevant stimuli across the cortical hierarchy, but the specific across-area interactions enabling stimulus selection are still unclear. Here, we propose that population gating, occurring within primary auditory cortex (A1) but controlled by top-down inputs from prelimbic region of medial prefrontal cortex (mPFC), can support across-area stimulus selection. Examining single-unit activity recorded while rats performed an auditory context-dependent task, we found that A1 encoded relevant and irrelevant stimuli along a common dimension of its neural space. Yet, the relevant stimulus encoding was enhanced along an extra dimension. In turn, mPFC encoded only the stimulus relevant to the ongoing context. To identify candidate mechanisms for stimulus selection within A1, we reverse-engineered low-rank RNNs trained on a similar task. Our analyses predicted that two context-modulated neural populations gated their preferred stimulus in opposite contexts, which we confirmed in further analyses of A1. Finally, we show in a two-region RNN how population gating within A1 could be controlled by top-down inputs from PFC, enabling flexible across-area communication despite fixed inter-areal connectivity.

Activity-dependent organization of prefrontal hub-networks for associative learning and signal transformation

Associative learning is crucial for adapting to environmental changes. Interactions among neuronal populations involving the dorso-medial prefrontal cortex (dmPFC) are proposed to regulate associative learning, but how these neuronal populations store and process information about the association remains unclear. Here we developed a pipeline for longitudinal two-photon imaging and computational dissection of neural population activities in male mouse dmPFC during fear-conditioning procedures, enabling us to detect learning-dependent changes in the dmPFC network topology. Using regularized regression methods and graphical modeling, we found that fear conditioning drove dmPFC reorganization to generate a neuronal ensemble encoding conditioned responses (CR) characterized by enhanced internal coactivity, functional connectivity, and association with conditioned stimuli (CS). Importantly, neurons strongly responding to unconditioned stimuli during conditioning subsequently became hubs of this novel associative network for the CS-to-CR transformation. Altogether, we demonstrate learning-dependent dynamic modulation of population coding structured on the activity-dependent formation of the hub network within the dmPFC.

Developmentally distinct architectures in top-down circuits

The medial prefrontal cortex (mPFC) plays a key role in learning, mood and decision making, including in how individuals respond to threats 1-6 . mPFC undergoes a uniquely protracted development, with changes in synapse density, cortical thickness, long-range connectivity, and neuronal encoding properties continuing into early adulthood 7-21 . Models suggest that before adulthood, the slow-developing mPFC cannot adequately regulate activity in faster-developing subcortical centers 22,23 . They propose that during development, the enhanced influence of subcortical systems underlies distinctive behavioural strategies of juveniles and adolescents and that increasing mPFC control over subcortical structures eventually allows adult behaviours to emerge. Yet it has remained unclear how a progressive strengthening of top-down control can lead to nonlinear changes in behaviour as individuals mature 24,25 . To address this discrepancy, here we monitored and manipulated activity in the developing brain as animals responded to threats, establishing direct causal links between frontolimbic circuit activity and the behavioural strategies of juvenile, adolescent and adult mice. Rather than a linear strengthening of mPFC synaptic connectivity progressively regulating behaviour, we uncovered multiple developmental switches in the behavioural roles of mPFC circuits targeting the basolateral amygdala (BLA) and nucleus accumbens (NAc). We show these changes are accompanied by axonal pruning coinciding with functional strengthening of synaptic connectivity in the mPFC-BLA and mPFC-NAc pathways, which mature at different rates. Our results reveal how developing mPFC circuits pass through distinct architectures that may make them optimally adapted to the demands of age-specific challenges.

Ventromedial prefrontal neurons represent self-states shaped by vicarious fear in male mice

Perception of fear induced by others in danger elicits complex vicarious fear responses and behavioral outputs. In rodents, observing a conspecific receive aversive stimuli leads to escape and freezing behavior. It remains unclear how these behavioral self-states in response to others in fear are neurophysiologically represented. Here, we assess such representations in the ventromedial prefrontal cortex (vmPFC), an essential site for empathy, in an observational fear (OF) paradigm in male mice. We classify the observer mouse's stereotypic behaviors during OF using a machine-learning approach. Optogenetic inhibition of the vmPFC specifically disrupts OF-induced escape behavior. In vivo Ca²⁺ imaging reveals that vmPFC neural populations represent intermingled information of other- and self-states. Distinct subpopulations are activated and suppressed by others' fear responses, simultaneously representing self-freezing states. This mixed selectivity requires inputs from the anterior cingulate cortex and the basolateral amygdala to regulate OF-induced escape behavior.

Plasticity of cortico-striatal neurons of the caudal anterior cingulate cortex during experimental neuropathic pain

Maladaptive neuronal plasticity is a main mechanism for the development and maintenance of pathological pain. Affective, motivational and cognitive deficits that are comorbid with pain involve cellular and synaptic modifications in the anterior cingulate cortex (ACC), a major brain mediator of pain perception. Here we use a model of neuropathic pain (NP) in male mice and ex-vivo electrophysiology to investigate whether layer 5 caudal ACC (cACC) neurons projecting to the dorsomedial striatum (DMS), a critical region for motivational regulation of behavior, are involved in aberrant neuronal plasticity. We found that while the intrinsic excitability of cortico-striatal cACC neurons (cACC-CS) was preserved in NP animals, excitatory postsynaptic potentials (EPSP) induced after stimulation of distal inputs were enlarged. The highest synaptic responses were evident both after single stimuli and in each of the EPSP that compose responses to trains of stimuli, and were accompanied by increased synaptically-driven action potentials. EPSP temporal summation was intact in ACC-CS neurons from NP mice, suggesting that the plastic changes were not due to alterations in dendritic integration but rather through synaptic mechanisms. These results demonstrate for the first time that NP affects cACC neurons that project to the DMS and reinforce the notion that maladaptive plasticity of the cortico-striatal pathway may be a key factor in sustaining pathological pain.

Persistence is driven by a prefrontal motor circuit

Persistence provides a long-lasting effect on actions, including avoiding predators and storing energy, and hence is crucial for the survival (Adolphs and Anderson, 2018). However, how the brain loads persistence on movements is unknown. Here, we demonstrate that being persistent is determined at the initial phase of movement, and this persistency will be sustained until the terminal signaling. The neural coding of persistent movement phases (initial or terminal) is independent from the judgement (i.e. valence) (Li et al., 2022; Wang et al., 2018) upon the external stimuli. Next, we identify a group of dorsal medial prefrontal cortex (dmPFC) motor cortex projecting (MP) neurons (Wang and Sun, 2021), which encodes the initial phase of a persistent movement rather than the valence. Inactivation of dmPFC MP neurons impairs the initiation of persistency and reduce the neural activity in the insular and motor cortex. Finally, a MP network-based computational model suggests that an intact, successive sensory stimulus acts as a triggering signal to direct the initiation of persistent movements. These findings reveal a neural mechanism that transforms the brain state from neutral to persistent during a movement.

Resolving the prefrontal mechanisms of adaptive cognitive behaviors: A cross-species perspective

The prefrontal cortex (PFC) enables a staggering variety of complex behaviors, such as planning actions, solving problems, and adapting to new situations according to external information and internal states. These higher-order abilities, collectively defined as adaptive cognitive behavior, require cellular ensembles that coordinate the tradeoff between the stability and flexibility of neural representations. While the mechanisms underlying the function of cellular ensembles are still unclear, recent experimental and theoretical studies suggest that temporal coordination dynamically binds prefrontal neurons into functional ensembles. A so far largely separate stream of research has investigated the prefrontal efferent and afferent connectivity. These two research streams have recently converged on the hypothesis that prefrontal connectivity patterns influence ensemble formation and the function of neurons within ensembles. Here, we propose a unitary concept that, leveraging a cross-species definition of prefrontal regions, explains how prefrontal ensembles adaptively regulate and efficiently coordinate multiple processes in distinct cognitive behaviors.

Insights into the encoding of memories through the circuitry of fear

Associative learning induces physical changes to a network of cells, known as the memory engram. Fear is widely used as a model to understand the circuit motifs that underpin associative memories. Recent advances suggest that the distinct circuitry engaged by different conditioned stimuli (e.g. tone vs. context) can provide insights into what information is being encoded in the fear engram. Moreover, as the fear memory matures, the circuitry engaged indicates how information is remodelled after learning and hints at potential mechanisms for consolidation. Finally, we propose that the consolidation of fear memories involves plasticity of engram cells through coordinated activity between brain regions, and the inherent characteristics of the circuitry may mediate this process.

Decoding defensive systems

Our understanding of the neuronal circuits and mechanisms of defensive systems has been primarily dominated by studies focusing on the contribution of individual cells in the processing of threat-predictive cues, defensive responses, the extinction of such responses and the contextual modulation of threat-related behavior. These studies have been key in establishing threat-related circuits and mechanisms. Yet, they fall short in answering long-standing questions related to the integrative processing of distinct threatening cues, behavioral states induced by threat-related events, or the bridging from sensory processing of threat-related cues to specific defensive responses. Recent conceptual and technical developments has allowed the monitoring of large populations of neurons, which in addition to advanced analytic tools, have improved our understanding of how collective neuronal activity supports threat-related behaviors. In this review, we discuss the current knowledge of neuronal population codes within threat-related networks, in the context of aversive motivated behavior and the study of defensive systems.

Neural Oscillations in Aversively Motivated Behavior

Fear and anxiety-based disorders are highly debilitating and among the most prevalent psychiatric disorders. These disorders are associated with abnormal network oscillations in the brain, yet a comprehensive understanding of the role of network oscillations in the regulation of aversively motivated behavior is lacking. In this review, we examine the oscillatory correlates of fear and anxiety with a particular focus on rhythms in the theta and gamma-range. First, we describe neural oscillations and their link to neural function by detailing the role of well-studied theta and gamma rhythms to spatial and memory functions of the hippocampus. We then describe how theta and gamma oscillations act to synchronize brain structures to guide adaptive fear and anxiety-like behavior. In short, that hippocampal network oscillations act to integrate spatial information with motivationally salient information from the amygdala during states of anxiety before routing this information via theta oscillations to appropriate target regions, such as the prefrontal cortex. Moreover, theta and gamma oscillations develop in the amygdala and neocortical areas during the encoding of fear memories, and interregional synchronization reflects the retrieval of both recent and remotely encoded fear memories. Finally, we argue that the thalamic nucleus reuniens represents a key node synchronizing prefrontal-hippocampal theta dynamics for the retrieval of episodic extinction memories in the hippocampus.

Divergent encoding of active avoidance behavior in corticostriatal and corticolimbic projections

Active avoidance behavior, in which an animal performs an action to avoid a stressor, is crucial for survival and may provide insight into avoidance behaviors seen in anxiety disorders. Active avoidance requires the dorsomedial prefrontal cortex (dmPFC), which is thought to regulate avoidance via downstream projections to the striatum and amygdala. However, the endogenous activity of dmPFC projections during active avoidance learning has never been recorded. Here we utilized fiber photometry to record from the dmPFC and its axonal projections to the dorsomedial striatum (DMS) and the basolateral amygdala (BLA) during active avoidance learning in both male and female mice. We examined neural activity during conditioned stimulus (CS) presentations and during clinically relevant behaviors such as active avoidance or cued freezing. Both prefrontal projections showed learning-related increases in activity during CS onset throughout active avoidance training. The dmPFC as a whole showed increased and decreased patterns of activity during avoidance and cued freezing, respectively. Finally, dmPFC-DMS and dmPFC-BLA projections show divergent encoding of active avoidance behavior, with the dmPFC-DMS projection showing increased activity and the dmPFC-BLA projection showing decreased activity during active avoidance. Our results demonstrate task-relevant encoding of active avoidance in projection-specific dmPFC subpopulations that play distinct but complementary roles in active avoidance learning.

Discriminating Sleep From Freezing With Cortical Spindle Oscillations

In-vivo longitudinal recordings require reliable means to automatically discriminate between distinct behavioral states, in particular between awake and sleep epochs. The typical approach is to use some measure of motor activity together with extracellular electrophysiological signals, namely the relative contribution of theta and delta frequency bands to the Local Field Potential (LFP). However, these bands can partially overlap with oscillations characterizing other behaviors such as the 4 Hz accompanying rodent freezing. Here, we first demonstrate how standard methods fail to discriminate between sleep and freezing in protocols where both behaviors are observed. Then, as an alternative, we propose to use the smoothed cortical spindle power to detect sleep epochs. Finally, we show the effectiveness of this method in discriminating between sleep and freezing in our recordings.

BehaviorDEPOT is a simple, flexible tool for automated behavioral detection based on markerless pose tracking

Quantitative descriptions of animal behavior are essential to study the neural substrates of cognitive and emotional processes. Analyses of naturalistic behaviors are often performed by hand or with expensive, inflexible commercial software. Recently, machine learning methods for markerless pose estimation enabled automated tracking of freely moving animals, including in labs with limited coding expertise. However, classifying specific behaviors based on pose data requires additional computational analyses and remains a significant challenge for many groups. We developed BehaviorDEPOT (DEcoding behavior based on POsitional Tracking), a simple, flexible software program that can detect behavior from video timeseries and can analyze the results of experimental assays. BehaviorDEPOT calculates kinematic and postural statistics from keypoint tracking data and creates heuristics that reliably detect behaviors. It requires no programming experience and is applicable to a wide range of behaviors and experimental designs. We provide several hard-coded heuristics. Our freezing detection heuristic achieves above 90% accuracy in videos of mice and rats, including those wearing tethered head-mounts. BehaviorDEPOT also helps researchers develop their own heuristics and incorporate them into the software's graphical interface. Behavioral data is stored framewise for easy alignment with neural data. We demonstrate the immediate utility and flexibility of BehaviorDEPOT using popular assays including fear conditioning, decision-making in a T-maze, open field, elevated plus maze, and novel object exploration.