Reward Association Enhances Stimulus-Specific Representations in Primary Visual Cortex

HIV-1 Tat protein alters medial prefrontal cortex neuronal activity and recognition memory

Despite advancements in combined antiretroviral therapy, human immunodeficiency virus (HIV)-associated neurocognitive disorder (HAND) continue to affect 40%-50% of people living with HIV. While neuroimaging studies have revealed HIV-1-induced alterations in cortical networks and brain macrostructures, it still remains unclear how individual neurons in the medial prefrontal cortex (mPFC) are affected during recognition memory. Using in vivo calcium imaging in an HIV-1 transactivator of transcription (Tat) transgenic mouse model, we examined mPFC neuronal activity during a novel object recognition memory task. Our findings show that HIV Tat expression reduces overall neuronal activity in Tat(+) mice without altering the number of activated cells. Moreover, distinct neuronal subpopulations are up- and downmodulated in both Tat(-) and Tat(+) mice depending on object exploration. Importantly, familiarity-driven increases in mPFC activity were disrupted by HIV Tat expression. These findings enhance our understanding of HAND and may inform future pharmacological strategies aimed at restoring cognitive function.

Amygdalo-cortical dialogue underlies memory enhancement by emotional association

Emotional arousal plays a critical role in determining what is remembered from experiences. It is hypothesized that activation of the amygdala by emotional stimuli enhances memory consolidation in its downstream brain regions. However, the physiological basis of the inter-regional interaction and its functions remain unclear. Here, by adding emotional information to a perceptual recognition task that relied on a frontal-sensory cortical circuit in mice, we demonstrated that the amygdala not only associates emotional information with perceptual information but also enhances perceptual memory retention via amygdalo-frontal cortical projections. Furthermore, emotional association increased reactivation of coordinated activity across the amygdalo-cortical circuit during non-rapid eye movement (NREM) sleep but not during rapid eye movement (REM) sleep. Notably, this increased reactivation was associated with amygdala high-frequency oscillations. Silencing of amygdalo-cortical inputs during NREM sleep selectively disrupted perceptual memory enhancement. Our findings indicate that inter-regional reactivation triggered by the amygdala during NREM sleep underlies emotion-induced perceptual memory enhancement.

Visual experience orthogonalizes visual cortical stimulus responses via population code transformation

Sensory and behavioral experience can alter visual cortical stimulus coding, but the precise form of this plasticity is unclear. We measured orientation tuning in 4,000-neuron populations of mouse V1 before and after training on a visuomotor task. Changes to single-cell tuning curves appeared complex, including development of asymmetries and of multiple peaks. Nevertheless, these complex tuning curve transformations can be explained by a simple equation: a convex transformation suppressing responses to task stimuli specifically in cells responding at intermediate levels. The strength of the transformation varies across trials, suggesting a dynamic circuit mechanism rather than static synaptic plasticity. The transformation results in sparsening and orthogonalization of population codes for task stimuli. It cannot improve the performance of an optimal stimulus decoder, which is already perfect even for naive codes, but it improves the performance of a suboptimal decoder model with inductive bias as might be found in downstream readout circuits.

Moculus: an immersive virtual reality system for mice incorporating stereo vision

Due to technical roadblocks, it is unclear how visual circuits represent multiple features or how behaviorally relevant representations are selected for long-term memory. Here we developed Moculus, a head-mounted virtual reality platform for mice that covers the entire visual field, and allows binocular depth perception and full visual immersion. This controllable environment, with three-dimensional (3D) corridors and 3D objects, in combination with 3D acousto-optical imaging, affords rapid visual learning and the uncovering of circuit substrates in one measurement session. Both the control and reinforcement-associated visual cue coding neuronal assemblies are transiently expanded by reinforcement feedback to near-saturation levels. This increases computational capability and allows competition among assemblies that encode behaviorally relevant information. The coding assemblies form partially orthogonal and overlapping clusters centered around hub cells with higher and earlier ramp-like responses, as well as locally increased functional connectivity.

Stability of cross-sensory input to primary somatosensory cortex across experience

Merging information across sensory modalities is key to forming robust percepts, yet how the brain achieves this feat remains unclear. Recent studies report cross-modal influences in the primary sensory cortex, suggesting possible multisensory integration in the early stages of cortical processing. We test several hypotheses about the function of auditory influences on mouse primary somatosensory cortex (S1) using in vivo two-photon calcium imaging. We found sound-evoked spiking activity in an extremely small fraction of cells, and this sparse activity did not encode auditory stimulus identity. Moreover, S1 did not encode information about specific audio-tactile feature conjunctions. Auditory and audio-tactile stimulus encoding remained unchanged after both passive experience and reinforcement. These results suggest that while primary sensory cortex is plastic within its own modality, the influence of other modalities is remarkably stable and stimulus nonspecific.

Discretized representations in V1 predict suboptimal orientation discrimination

Neuronal population activity in sensory cortices is the substrate for perceptual decisions. Yet, we still do not understand how neuronal information content in sensory cortices relates to behavioral reports. To reconcile neurometric and psychometric performance, we recorded the activity of V1 neurons in mice performing a Go/NoGo orientation discrimination task. We found that, around the discrimination threshold, V1 does not represent the orientation of the stimuli as canonically expected. Instead, it forms categorical representations characterized by a relocation of activity at task-relevant domains of the orientation representational space. The relative neuronal activity at those discrete domains accurately predicted the probabilities of the animals' decisions. Our results thus suggest that the categorical integration of discretized feature representations from sensory cortices explains perceptual decisions.

Learning enhances behaviorally relevant representations in apical dendrites

Learning alters cortical representations and improves perception. Apical tuft dendrites in cortical layer 1, which are unique in their connectivity and biophysical properties, may be a key site of learning-induced plasticity. We used both two-photon and SCAPE microscopy to longitudinally track tuft-wide calcium spikes in apical dendrites of layer 5 pyramidal neurons in barrel cortex as mice learned a tactile behavior. Mice were trained to discriminate two orthogonal directions of whisker stimulation. Reinforcement learning, but not repeated stimulus exposure, enhanced tuft selectivity for both directions equally, even though only one was associated with reward. Selective tufts emerged from initially unresponsive or low-selectivity populations. Animal movement and choice did not account for changes in stimulus selectivity. Enhanced selectivity persisted even after rewards were removed and animals ceased performing the task. We conclude that learning produces long-lasting realignment of apical dendrite tuft responses to behaviorally relevant dimensions of a task.

Adolescent-like Processing of Behaviorally Salient Cues in Sensory and Prefrontal Cortices of Adult Preterm-Born Mice

Preterm birth is a leading risk factor for atypicalities in cognitive and sensory processing, but it is unclear how prematurity impacts circuits that support these functions. To address this, we trained adult mice born a day early (preterm mice) on a visual discrimination task and found that they commit more errors and fail to achieve high levels of performance. Using in vivo electrophysiology, we found that the neurons in the primary visual cortex (V1) and the V1-projecting prefrontal anterior cingulate cortex (ACC) are hyper-responsive to the reward, reminiscent of cue processing in adolescence. Moreover, the non-rewarded cue fails to robustly activate the V1 and V1-projecting ACC neurons during error trials, in contrast to prefrontal fast-spiking (FS) interneurons which show elevated error-related activity, suggesting that preterm birth impairs the function of prefrontal circuits for error monitoring. Finally, environmental enrichment, a well-established paradigm that promotes sensory maturation, failed to improve the performance of preterm mice, suggesting limited capacity of early interventions for reducing the risk of cognitive deficits after preterm birth. Altogether, our study for the first time identifies potential circuit mechanisms of cognitive atypicalities in the preterm population and highlights the vulnerability of prefrontal circuits to advanced onset of extrauterine experience.

Adolescent-like Processing of Behaviorally Salient Cues in Sensory and Prefrontal Cortices of Adult Preterm-Born Mice

Preterm birth is a leading risk factor for atypicalities in cognitive and sensory processing, but it is unclear how prematurity impacts circuits that support these functions. To address this, we trained adult mice born a day early (preterm mice) on a visual discrimination task and found that they commit more errors and fail to achieve high levels of performance. Using in vivo electrophysiology , we found that the neurons in the primary visual cortex (V1) and the V1-projecting prefrontal anterior cingulate cortex (ACC) are hyper-responsive to the reward, reminiscent of cue processing in adolescence. Moreover, the non-rewarded cue fails to robustly activate the V1 and V1-projecting ACC neurons during error trials, in contrast to prefrontal fast-spiking (FS) interneurons which show elevated error-related activity, suggesting that preterm birth impairs the function of prefrontal circuits for error monitoring. Finally, environmental enrichment, a well-established paradigm that promotes sensory maturation, failed to improve the performance of preterm mice, suggesting limited capacity of early interventions for reducing the risk of cognitive deficits after preterm birth. Altogether, our study for the first time identifies potential circuit mechanisms of cognitive atypicalities in the preterm population and highlights the vulnerability of prefrontal circuits to advanced onset of extrauterine experience.

Long-lasting, subtype-specific regulation of somatostatin interneurons during sensory learning

Somatostatin (SST)-expressing inhibitory neurons are a major class of neocortical γ-amino butyric acid (GABA) neurons, where morphological, electrophysiological, and transcriptomic analyses indicate more than a dozen different subtypes. However, whether this diversity is related to specific roles in cortical computations and plasticity remains unclear. Here we identify learning-dependent, subtype-specific plasticity in layer 2/3 SST neurons of the mouse somatosensory cortex. Martinotti-type, SST neurons expressing calbindin-2 show a selective decrease in excitatory synaptic input and stimulus-evoked calcium responses as mice learn a stimulus-reward association. Using these insights, we develop a label-free classifier using basal activity from in vivo imaging that accurately predicts learning-associated response plasticity. Our data indicate that molecularly-defined SST neuron subtypes play specific and highly-regulated roles in sensory information processing and learning.

Stimulus repetition induces a two-stage learning process in primary visual cortex

Repeated stimulus exposure alters the brain's response to the stimulus. We investigated the underlying neural mechanisms by recording functional MRI data from human observers passively viewing 120 presentations of two Gabor patches (each Gabor repeating 60 times). We evaluated support for two prominent models of stimulus repetition, the fatigue model and the sharpening model. Our results uncovered a two-stage learning process in the primary visual cortex. In Stage 1, univariate BOLD activation in V1 decreased over the first twelve repetitions of the stimuli, replicating the well-known effect of repetition suppression. Applying MVPA decoding along with a moving window approach, we found that (1) the decoding accuracy between the two Gabors decreased from above-chance level (∼60% to ∼70%) at the beginning of the stage to chance level at the end of the stage (∼50%). This result, together with the accompanying weight map analysis, suggested that the learning dynamics in Stage 1 were consistent with the predictions of the fatigue model. In Stage 2, univariate BOLD activation for the remaining 48 repetitions of the two stimuli exhibited significant fluctuations but no systematic trend. The MVPA decoding accuracy between the two Gabor patches was at chance level initially and became progressively higher as stimulus repetition continued, rising above and staying above chance level starting at the ∼35th repetition. Thus, results from the second stage supported the notion that sustained and prolonged stimulus repetition prompts sharpened representations. Additional analyses addressed (1) whether the neural patterns within each learning stage remained stable and (2) whether new neural patterns were evoked in Stage 2 relative to Stage 1.

Glucagon-like peptide 1 agonist and effects on reward behaviour: A systematic review

INTRODUCTION

The roles of metabolic signals, including Glucagon-like peptide 1 (GLP-1), have been implicated in multiple domains outside metabolic regulation. There is a growing interest in repurposing Glucagon-like peptide 1 receptor agonists (GLP-1RAs) as therapeutics for motivation and reward-related behavioural disturbances. Herein, we aim to systematically review the extant evidence on the potential effects of GLP-1RAs on the reward system.

METHODS

The study followed PRISMA guidelines using databases such as OVID, PubMed, Scopus, and Google Scholar. The search focused on "Reward Behavior" and "Glucagon Like Peptide 1 Receptor Agonists" and was restricted to human studies. Quality assessment achieved by the NIH's Quality Assessment of Controlled Intervention Studies RESULTS: GLP-1RAs consistently reduced energy intake and influenced reward-related behaviour. These agents have been associated with decreased neurocortical activation in response to higher rewards and food cues, particularly high-calorie foods, and lowered caloric intake and hunger levels.

DISCUSSION

GLP-1RAs show promise in addressing reward dysfunction linked to food stimuli, obesity, and T2DM. They normalize insulin resistance, and might also modulate dopaminergic signalling and reduce anhedonia. Their effects on glycemic variability and cravings suggest potential applications in addiction disorders.

Stability of cross-sensory input to primary somatosensory cortex across experience

Merging information from across sensory modalities is key to forming robust, disambiguated percepts of the world, yet how the brain achieves this feat remains unclear. Recent observations of cross-modal influences in primary sensory cortical areas have suggested that multisensory integration may occur in the earliest stages of cortical processing, but the role of these responses is still poorly understood. We address these questions by testing several hypotheses about the possible functions served by auditory influences on the barrel field of mouse primary somatosensory cortex (S1) using in vivo 2-photon calcium imaging. We observed sound-evoked spiking activity in a small fraction of cells overall, and moreover that this sparse activity was insufficient to encode auditory stimulus identity; few cells responded preferentially to one sound or another, and a linear classifier trained to decode auditory stimuli from population activity performed barely above chance. Moreover S1 did not encode information about specific audio-tactile feature conjunctions that we tested. Our ability to decode auditory audio-tactile stimuli from neural activity remained unchanged after both passive experience and reinforcement. Collectively, these results suggest that while a primary sensory cortex is highly plastic with regard to its own modality, the influence of other modalities are remarkably stable and play a largely stimulus-non-specific role.

Assessment of visual function in blind mice and monkeys with subretinally implanted nanowire arrays as artificial photoreceptors

Retinal prostheses could restore image-forming vision in conditions of photoreceptor degeneration. However, contrast sensitivity and visual acuity are often insufficient. Here we report the performance, in mice and monkeys with induced photoreceptor degeneration, of subretinally implanted gold-nanoparticle-coated titania nanowire arrays providing a spatial resolution of 77.5 μm and a temporal resolution of 3.92 Hz in ex vivo retinas (as determined by patch-clamp recording of retinal ganglion cells). In blind mice, the arrays allowed for the detection of drifting gratings and flashing objects at light-intensity thresholds of 15.70-18.09 μW mm-2, and offered visual acuities of 0.3-0.4 cycles per degree, as determined by recordings of visually evoked potentials and optomotor-response tests. In monkeys, the arrays were stable for 54 weeks, allowed for the detection of a 10-μW mm-2 beam of light (0.5° in beam angle) in visually guided saccade experiments, and induced plastic changes in the primary visual cortex, as indicated by long-term in vivo calcium imaging. Nanomaterials as artificial photoreceptors may ameliorate visual deficits in patients with photoreceptor degeneration.

Beta activity in human anterior cingulate cortex mediates reward biases

The rewards that we get from our choices and actions can have a major influence on our future behavior. Understanding how reward biasing of behavior is implemented in the brain is important for many reasons, including the fact that diminution in reward biasing is a hallmark of clinical depression. We hypothesized that reward biasing is mediated by the anterior cingulate cortex (ACC), a cortical hub region associated with the integration of reward and executive control and with the etiology of depression. To test this hypothesis, we recorded neural activity during a biased judgment task in patients undergoing intracranial monitoring for either epilepsy or major depressive disorder. We found that beta (12-30 Hz) oscillations in the ACC predicted both associated reward and the size of the choice bias, and also tracked reward receipt, thereby predicting bias on future trials. We found reduced magnitude of bias in depressed patients, in whom the beta-specific effects were correspondingly reduced. Our findings suggest that ACC beta oscillations may orchestrate the learning of reward information to guide adaptive choice, and, more broadly, suggest a potential biomarker for anhedonia and point to future development of interventions to enhance reward impact for therapeutic benefit.

Sex-specific resilience of neocortex to food restriction

Mammals have evolved sex-specific adaptations to reduce energy usage in times of food scarcity. These adaptations are well described for peripheral tissue, though much less is known about how the energy-expensive brain adapts to food restriction, and how such adaptations differ across the sexes. Here, we examined how food restriction impacts energy usage and function in the primary visual cortex (V1) of adult male and female mice. Molecular analysis and RNA sequencing in V1 revealed that in males, but not in females, food restriction significantly modulated canonical, energy-regulating pathways, including pathways associated waith AMP-activated protein kinase, peroxisome proliferator-activated receptor alpha, mammalian target of rapamycin, and oxidative phosphorylation. Moreover, we found that in contrast to males, food restriction in females did not significantly affect V1 ATP usage or visual coding precision (assessed by orientation selectivity). Decreased serum leptin is known to be necessary for triggering energy-saving changes in V1 during food restriction. Consistent with this, we found significantly decreased serum leptin in food-restricted males but no significant change in food-restricted females. Collectively, our findings demonstrate that cortical function and energy usage in female mice are more resilient to food restriction than in males. The neocortex, therefore, contributes to sex-specific, energy-saving adaptations in response to food restriction.

Transient enhancement of stimulus-evoked activity in neocortex during sensory learning

Synaptic potentiation has been linked to learning in sensory cortex, but the connection between this potentiation and increased sensory-evoked neural activity is not clear. Here, we used longitudinal in vivo Ca2+ imaging in the barrel cortex of awake mice to test the hypothesis that increased excitatory synaptic strength during the learning of a whisker-dependent sensory-association task would be correlated with enhanced stimulus-evoked firing. To isolate stimulus-evoked responses from dynamic, task-related activity, imaging was performed outside of the training context. Although prior studies indicate that multiwhisker stimuli drive robust subthreshold activity, we observed sparse activation of L2/3 pyramidal (Pyr) neurons in both control and trained mice. Despite evidence for excitatory synaptic strengthening at thalamocortical and intracortical synapses in this brain area at the onset of learning-indeed, under our imaging conditions thalamocortical axons were robustly activated-we observed that L2/3 Pyr neurons in somatosensory (barrel) cortex displayed only modest increases in stimulus-evoked activity that were concentrated at the onset of training. Activity renormalized over longer training periods. In contrast, when stimuli and rewards were uncoupled in a pseudotraining paradigm, stimulus-evoked activity in L2/3 Pyr neurons was significantly suppressed. These findings indicate that sensory-association training but not sensory stimulation without coupled rewards may briefly enhance sensory-evoked activity, a phenomenon that might help link sensory input to behavioral outcomes at the onset of learning.

Trait food craving predicts functional connectivity between dopaminergic midbrain and the fusiform food area during eating imagery

Neurofunctional coupling between the dopaminergic midbrain (i.e., ventral tegmental area, VTA) and higher-order visual regions may contribute to food craving, leading to the onset or maintenance of obesity. We recently showed that the VTA resting-state functional connectivity with the occipitotemporal cortex, at the level of the fusiform gyrus (FFG), was specifically associated with trait food craving and the implicit bias for food images, suggesting that VTA-FFG connectivity may reflect the association between the visual representations of food and its motivational properties. To further test this hypothesis, this time we studied task-based functional connectivity in twenty-eight healthy-weight participants while imagining eating their most liked high-calorie (HC) or least liked low-calorie food (LC) or drinking water (control condition). Trait food craving scores were used to predict changes in task-based functional connectivity of the VTA during imagery of HC compared to LC foods (relative to the control condition). Trait food craving was positively associated with the functional connectivity of the VTA with the left FFG: people with higher trait food craving scores show stronger VTA-FFG connectivity, specifically for the imagery of the liked HC foods. This association was not linked to the quality of imagery nor to state measures of craving, appetite, or thirst. These findings emphasize the contribution of the functional coupling between dopaminergic midbrain and higher-order visual regions to food craving, suggesting a neurofunctional mechanism by which the mental representations of the HC food we like can become much more salient if not irresistible.

Reward Modulates Visual Responses in the Superficial Superior Colliculus of Mice

The processing of sensory input is constantly adapting to behavioral demands and internal states. The drive to obtain reward, e.g., searching for water when thirsty, is a strong behavioral demand and associating the reward with its source, a certain environment or action, is paramount for survival. Here, we show that water reward increases subsequent visual activity in the superficial layers of the superior colliculus (SC), which receive direct input from the retina and belong to the earliest stages of visual processing. We trained mice of either sex to perform a visual decision task and recorded the activity of neurons in the SC using two-photon calcium imaging and high-density electrophysiological recordings. Responses to visual stimuli in around 20% of visually responsive neurons in the superficial SC were affected by reward delivered in the previous trial. Reward mostly increased visual responses independent from modulations due to pupil size changes. The modulation of visual responses by reward could not be explained by movements like licking. It was specific to responses to the following visual stimulus, independent of slow fluctuations in neural activity and independent of how often the stimulus was previously rewarded. Electrophysiological recordings confirmed these results and revealed that reward affected the early phase of the visual response around 80 ms after stimulus onset. Modulation of visual responses by reward, but not pupil size, significantly improved the performance of a population decoder to detect visual stimuli, indicating the relevance of reward modulation for the visual performance of the animal.SIGNIFICANCE STATEMENT To learn which actions lead to food, water, or safety, it is necessary to integrate the receiving of reward with sensory stimuli related to the reward. Cortical stages of sensory processing have been shown to represent stimulus-reward associations. Here, we show, however, that reward influences neurons at a much earlier stage of sensory processing, the superior colliculus (SC), receiving direct input from the retina. Visual responses were increased shortly after the animal received the water reward, which led to an improved stimulus signal in the population of these visual neurons. Reward modulation of early visual responses may thus improve perception of visual environments predictive of reward.

Signatures of task learning in neural representations

While neural plasticity has long been studied as the basis of learning, the growth of large-scale neural recording techniques provides a unique opportunity to study how learning-induced activity changes are coordinated across neurons within the same circuit. These distributed changes can be understood through an evolution of the geometry of neural manifolds and latent dynamics underlying new computations. In parallel, studies of multi-task and continual learning in artificial neural networks hint at a tradeoff between non-interference and compositionality as guiding principles to understand how neural circuits flexibly support multiple behaviors. In this review, we highlight recent findings from both biological and artificial circuits that together form a new framework for understanding task learning at the population level.

Repeated passive visual experience modulates spontaneous and non-familiar stimuli-evoked neural activity

Familiarity creates subjective memory of repeated innocuous experiences, reduces neural and behavioral responsiveness to those experiences, and enhances novelty detection. The neural correlates of the internal model of familiarity and the cellular mechanisms of enhanced novelty detection following multi-day repeated passive experience remain elusive. Using the mouse visual cortex as a model system, we test how the repeated passive experience of a 45° orientation-grating stimulus for multiple days alters spontaneous and non-familiar stimuli evoked neural activity in neurons tuned to familiar or non-familiar stimuli. We found that familiarity elicits stimulus competition such that stimulus selectivity reduces in neurons tuned to the familiar 45° stimulus; it increases in those tuned to the 90° stimulus but does not affect neurons tuned to the orthogonal 135° stimulus. Furthermore, neurons tuned to orientations 45° apart from the familiar stimulus dominate local functional connectivity. Interestingly, responsiveness to natural images, which consists of familiar and non-familiar orientations, increases subtly in neurons that exhibit stimulus competition. We also show the similarity between familiar grating stimulus-evoked and spontaneous activity increases, indicative of an internal model of altered experience.

Propagation of activity through the cortical hierarchy and perception are determined by neural variability

Brains are composed of anatomically and functionally distinct regions performing specialized tasks, but regions do not operate in isolation. Orchestration of complex behaviors requires communication between brain regions, but how neural dynamics are organized to facilitate reliable transmission is not well understood. Here we studied this process directly by generating neural activity that propagates between brain regions and drives behavior, assessing how neural populations in sensory cortex cooperate to transmit information. We achieved this by imaging two densely interconnected regions-the primary and secondary somatosensory cortex (S1 and S2)-in mice while performing two-photon photostimulation of S1 neurons and assigning behavioral salience to the photostimulation. We found that the probability of perception is determined not only by the strength of the photostimulation but also by the variability of S1 neural activity. Therefore, maximizing the signal-to-noise ratio of the stimulus representation in cortex relative to the noise or variability is critical to facilitate activity propagation and perception.

Repeated passive visual experience modulates spontaneous and non-familiar stimulievoked neural activity

Familiarity creates subjective memory of repeated innocuous experiences, reduces neural and behavioral responsiveness to those experiences, and enhances novelty detection. The neural correlates of the internal model of familiarity and the cellular mechanisms of enhanced novelty detection following multi-day repeated passive experience remain elusive. Using the mouse visual cortex as a model system, we test how the repeated passive experience of a 45° orientation-grating stimulus for multiple days alters spontaneous and non-familiar stimuli evoked neural activity in neurons tuned to familiar or non-familiar stimuli. We found that familiarity elicits stimulus competition such that stimulus selectivity reduces in neurons tuned to the familiar 45° stimulus; it increases in those tuned to the 90° stimulus but does not affect neurons tuned to the orthogonal 135° stimulus. Furthermore, neurons tuned to orientations 45° apart from the familiar stimulus dominate local functional connectivity. Interestingly, responsiveness to natural images, which consists of familiar and non-familiar orientations, increases subtly in neurons that exhibit stimulus competition. We also show the similarity between familiar grating stimulus-evoked and spontaneous activity increases, indicative of an internal model of altered experience.

Discovering Low-Dimensional Descriptions of Multineuronal Dependencies

Coordinated activity in neural populations is crucial for information processing. Shedding light on the multivariate dependencies that shape multineuronal responses is important to understand neural codes. However, existing approaches based on pairwise linear correlations are inadequate at capturing complicated interaction patterns and miss features that shape aspects of the population function. Copula-based approaches address these shortcomings by extracting the dependence structures in the joint probability distribution of population responses. In this study, we aimed to dissect neural dependencies with a C-Vine copula approach coupled with normalizing flows for estimating copula densities. While this approach allows for more flexibility compared to fitting parametric copulas, drawing insights on the significance of these dependencies from large sets of copula densities is challenging. To alleviate this challenge, we used a weighted non-negative matrix factorization procedure to leverage shared latent features in neural population dependencies. We validated the method on simulated data and applied it on copulas we extracted from recordings of neurons in the mouse visual cortex as well as in the macaque motor cortex. Our findings reveal that neural dependencies occupy low-dimensional subspaces, but distinct modules are synergistically combined to give rise to diverse interaction patterns that may serve the population function.

Emerging V1 neuronal ensembles with enhanced connectivity after associative learning

Introduction

The visual stimulus-specific responses in the primary visual cortex (V1) undergo plastic changes after associative learning. During the learning process, neuronal ensembles, defined as groups of coactive neurons, are well known to be related to learning and memory. However, it remains unclear what effect learning has on ensembles, and which neuronal subgroups within those ensembles play a key role in associative learning.

Methods

We used two-photon calcium imaging in mice to record the activity of V1 neurons before and after fear conditioning associated with a visual cue (blue light). We first defined neuronal ensembles by thresholding their functional connectivity in response to blue (conditioned) or green (control) light. We defined neurons that existed both before and after conditioning as stable neurons. Neurons which were recruited after conditioning were defined as new neurons. The graph theory-based analysis was performed to quantify the changes in connectivity within ensembles after conditioning.

Results

A significant enhancement in the connectivity strength (the average correlation with other neurons) was observed in the blue ensembles after conditioning. We found that stable neurons within the blue ensembles showed a significantly smaller clustering coefficient (the value represented the degree of interconnectedness among a node's neighbors) after conditioning than they were before conditioning. Additionally, new neurons within the blue ensembles had a larger clustering coefficient, similar relative degree (the value represented the number of functional connections between neurons) and connectivity strength compared to stable neurons in the same ensembles.

Discussion

Overall, our results demonstrated that the plastic changes caused by conditioning occurred in subgroups of neurons in the ensembles. Moreover, new neurons from conditioned ensembles may play a crucial role in memory formation, as they exhibited not only similar connection competence in relative degree and connectivity strength as stable neurons, but also showed a significantly larger clustering coefficient compared to the stable neurons within the same ensembles after conditioning.

Expectation violations produce error signals in mouse V1

Repeated exposure to visual sequences changes the form of evoked activity in the primary visual cortex (V1). Predictive coding theory provides a potential explanation for this, namely that plasticity shapes cortical circuits to encode spatiotemporal predictions and that subsequent responses are modulated by the degree to which actual inputs match these expectations. Here we use a recently developed statistical modeling technique called Model-Based Targeted Dimensionality Reduction (MbTDR) to study visually evoked dynamics in mouse V1 in the context of an experimental paradigm called "sequence learning." We report that evoked spiking activity changed significantly with training, in a manner generally consistent with the predictive coding framework. Neural responses to expected stimuli were suppressed in a late window (100-150 ms) after stimulus onset following training, whereas responses to novel stimuli were not. Substituting a novel stimulus for a familiar one led to increases in firing that persisted for at least 300 ms. Omitting predictable stimuli in trained animals also led to increased firing at the expected time of stimulus onset. Finally, we show that spiking data can be used to accurately decode time within the sequence. Our findings are consistent with the idea that plasticity in early visual circuits is involved in coding spatiotemporal information.

Optogenetic manipulation of inhibitory interneurons can be used to validate a model of spatiotemporal sequence learning

The brain uses temporal information to link discrete events into memory structures supporting recognition, prediction, and a wide variety of complex behaviors. It is still an open question how experience-dependent synaptic plasticity creates memories including temporal and ordinal information. Various models have been proposed to explain how this could work, but these are often difficult to validate in a living brain. A recent model developed to explain sequence learning in the visual cortex encodes intervals in recurrent excitatory synapses and uses a learned offset between excitation and inhibition to generate precisely timed "messenger" cells that signal the end of an instance of time. This mechanism suggests that the recall of stored temporal intervals should be particularly sensitive to the activity of inhibitory interneurons that can be easily targeted in vivo with standard optogenetic tools. In this work we examined how simulated optogenetic manipulations of inhibitory cells modifies temporal learning and recall based on these mechanisms. We show that disinhibition and excess inhibition during learning or testing cause characteristic errors in recalled timing that could be used to validate the model in vivo using either physiological or behavioral measurements.

Tasks and their role in visual neuroscience

Vision is widely used as a model system to gain insights into how sensory inputs are processed and interpreted by the brain. Historically, careful quantification and control of visual stimuli have served as the backbone of visual neuroscience. There has been less emphasis, however, on how an observer's task influences the processing of sensory inputs. Motivated by diverse observations of task-dependent activity in the visual system, we propose a framework for thinking about tasks, their role in sensory processing, and how we might formally incorporate tasks into our models of vision.

Amplified cortical neural responses as animals learn to use novel activity patterns

Cerebral cortex supports representations of the world in patterns of neural activity, used by the brain to make decisions and guide behavior. Past work has found diverse, or limited, changes in the primary sensory cortex in response to learning, suggesting that the key computations might occur in downstream regions. Alternatively, sensory cortical changes may be central to learning. We studied cortical learning by using controlled inputs we insert: we trained mice to recognize entirely novel, non-sensory patterns of cortical activity in the primary visual cortex (V1) created by optogenetic stimulation. As animals learned to use these novel patterns, we found that their detection abilities improved by an order of magnitude or more. The behavioral change was accompanied by large increases in V1 neural responses to fixed optogenetic input. Neural response amplification to novel optogenetic inputs had little effect on existing visual sensory responses. A recurrent cortical model shows that this amplification can be achieved by a small mean shift in recurrent network synaptic strength. Amplification would seem to be desirable to improve decision-making in a detection task; therefore, these results suggest that adult recurrent cortical plasticity plays a significant role in improving behavioral performance during learning.

Engaging distributed cortical and cerebellar networks through motor execution, observation, and imagery

When we interact with the environment around us, we are sometimes active participants, making directed physical motor movements and other times only mentally engaging with our environment, taking in sensory information and internally planning our next move without directed physical movement. Traditionally, cortical motor regions and key subcortical structures such as the cerebellum have been tightly linked to motor initiation, coordination, and directed motor behavior. However, recent neuroimaging studies have noted the activation of the cerebellum and wider cortical networks specifically during various forms of motor processing, including the observations of actions and mental rehearsal of movements through motor imagery. This phenomenon of cognitive engagement of traditional motor networks raises the question of how these brain regions are involved in the initiation of movement without physical motor output. Here, we will review evidence for distributed brain network activation during motor execution, observation, and imagery in human neuroimaging studies as well as the potential for cerebellar involvement specifically in motor-related cognition. Converging evidence suggests that a common global brain network is involved in both movement execution and motor observation or imagery, with specific task-dependent shifts in these global activation patterns. We will further discuss underlying cross-species anatomical support for these cognitive motor-related functions as well as the role of cerebrocerebellar communication during action observation and motor imagery.

Goals, usefulness and abstraction in value-based choice

Colombian drug lord Pablo Escobar, while on the run, purportedly burned two million dollars in banknotes to keep his daughter warm. A stark reminder that, in life, circumstances and goals can quickly change, forcing us to reassess and modify our values on-the-fly. Studies in decision-making and neuroeconomics have often implicitly equated value to reward, emphasising the hedonic and automatic aspect of the value computation, while overlooking its functional (concept-like) nature. Here we outline the computational and biological principles that enable the brain to compute the usefulness of an option or action by creating abstractions that flexibly adapt to changing goals. We present different algorithmic architectures, comparing ideas from artificial intelligence (AI) and cognitive neuroscience with psychological theories and, when possible, drawing parallels.

Cortical sensory processing across motivational states during goal-directed behavior

Behavioral states can influence performance of goal-directed sensorimotor tasks. Yet, it is unclear how altered neuronal sensory representations in these states relate to task performance and learning. We trained water-restricted mice in a two-whisker discrimination task to study cortical circuits underlying perceptual decision-making under different levels of thirst. We identified somatosensory cortices as well as the premotor cortex as part of the circuit necessary for task execution. Two-photon calcium imaging in these areas identified populations selective to sensory or motor events. Analysis of task performance during individual sessions revealed distinct behavioral states induced by decreasing levels of thirst-related motivation. Learning was better explained by improvements in motivational state control rather than sensorimotor association. Whisker sensory representations in the cortex were altered across behavioral states. In particular, whisker stimuli could be better decoded from neuronal activity during high task performance states, suggesting that state-dependent changes of sensory processing influence decision-making.

Neural function underlying reward expectancy and attainment in adolescents with diverse psychiatric symptoms

Reward dysfunction has been hypothesized to play a key role in the development of psychiatric conditions during adolescence. To help capture the complexity of reward function in youth, we used the Reward Flanker fMRI Task, which enabled us to examine neural activity during expectancy and attainment of both certain and uncertain rewards. Participants were 84 psychotropic-medication-free adolescents, including 67 with diverse psychiatric conditions and 17 healthy controls. Functional MRI used high-resolution acquisition and high-fidelity processing techniques modeled after the Human Connectome Project. Analyses examined neural activation during reward expectancy and attainment, and their associations with clinical measures of depression, anxiety, and anhedonia severity, with results controlled for family-wise errors using non-parametric permutation tests. As anticipated, reward expectancy activated regions within the fronto-striatal reward network, thalamus, occipital lobe, superior parietal lobule, temporoparietal junction, and cerebellum. Unexpectedly, however, reward attainment was marked by widespread deactivation in many of these same regions, which we further explored using cosine similarity analysis. Across all subjects, striatum and thalamus activation during reward expectancy negatively correlated with anxiety severity, while activation in numerous cortical and subcortical regions during reward attainment positively correlated with both anxiety and depression severity. These findings highlight the complexity and dynamic nature of neural reward processing in youth.

A flexible Python-based touchscreen chamber for operant conditioning reveals improved visual perception of cardinal orientations in mice

Natural scenes are composed of a wide range of edge angles and spatial frequencies, with a strong overrepresentation of vertical and horizontal edges. Correspondingly, many mammalian species are much better at discriminating these cardinal orientations compared to obliques. A potential reason for this increased performance could be an increased number of neurons in the visual cortex that are tuned to cardinal orientations, which is likely to be an adaptation to the natural scene statistics. Such biased angular tuning has recently been shown in the mouse primary visual cortex. However, it is still unknown if mice also show a perceptual dominance of cardinal orientations. Here, we describe the design of a novel custom-built touchscreen chamber that allows testing natural scene perception and orientation discrimination performance by applying different task designs. Using this chamber, we applied an iterative convergence towards orientation discrimination thresholds for cardinal or oblique orientations in different cohorts of mice. Surprisingly, the expert discrimination performance was similar for both groups but showed large inter-individual differences in performance and training time. To study the discrimination of cardinal and oblique stimuli in the same mice, we, therefore, applied, a different training regime where mice learned to discriminate cardinal and oblique gratings in parallel. Parallel training revealed a higher task performance for cardinal orientations in an early phase of the training. The performance for both orientations became similar after prolonged training, suggesting that learning permits equally high perceptual tuning towards oblique stimuli. In summary, our custom-built touchscreen chamber offers a flexible tool to test natural visual perception in rodents and revealed a training-induced increase in the perception of oblique gratings. The touchscreen chamber is entirely open-source, easy to build, and freely available to the scientific community to conduct visual or multimodal behavioral studies. It is also based on the FAIR principles for data management and sharing and could therefore serve as a catalyst for testing the perception of complex and natural visual stimuli across behavioral labs.

Mixed vine copula flows for flexible modeling of neural dependencies

Recordings of complex neural population responses provide a unique opportunity for advancing our understanding of neural information processing at multiple scales and improving performance of brain computer interfaces. However, most existing analytical techniques fall short of capturing the complexity of interactions within the concerted population activity. Vine copula-based approaches have shown to be successful at addressing complex high-order dependencies within the population, disentangled from the single-neuron statistics. However, most applications have focused on parametric copulas which bear the risk of misspecifying dependence structures. In order to avoid this risk, we adopted a fully non-parametric approach for the single-neuron margins and copulas by using Neural Spline Flows (NSF). We validated the NSF framework on simulated data of continuous and discrete types with various forms of dependency structures and with different dimensionality. Overall, NSFs performed similarly to existing non-parametric estimators, while allowing for considerably faster and more flexible sampling which also enables faster Monte Carlo estimation of copula entropy. Moreover, our framework was able to capture low and higher order heavy tail dependencies in neuronal responses recorded in the mouse primary visual cortex during a visual learning task while the animal was navigating a virtual reality environment. These findings highlight an often ignored aspect of complexity in coordinated neuronal activity which can be important for understanding and deciphering collective neural dynamics for neurotechnological applications.

Learning enhances encoding of time and temporal surprise in mouse primary sensory cortex

Primary sensory cortex has long been believed to play a straightforward role in the initial processing of sensory information. Yet, the superficial layers of cortex overall are sparsely active, even during sensory stimulation; additionally, cortical activity is influenced by other modalities, task context, reward, and behavioral state. Our study demonstrates that reinforcement learning dramatically alters representations among longitudinally imaged neurons in superficial layers of mouse primary somatosensory cortex. Learning an object detection task recruits previously unresponsive neurons, enlarging the neuronal population sensitive to touch and behavioral choice. Cortical responses decrease upon repeated stimulus presentation outside of the behavioral task. Moreover, training improves population encoding of the passage of time, and unexpected deviations in trial timing elicit even stronger responses than touches do. In conclusion, the superficial layers of sensory cortex exhibit a high degree of learning-dependent plasticity and are strongly modulated by non-sensory but behaviorally-relevant features, such as timing and surprise.

Activity in the superficial layers of the sensory cortex is believed to be largely driven by incoming sensory stimuli. Here the authors demonstrate how learning changes neural responses to sensations according to both behavioral relevance and timing, suggesting a high degree of non-sensory modulation.

Selective enhancement of neural coding in V1 underlies fine-discrimination learning in tree shrew

Visual discrimination improves with training, a phenomenon that is thought to reflect plastic changes in the responses of neurons in primary visual cortex (V1). However, the identity of the neurons that undergo change, the nature of the changes, and the consequences of these changes for other visual behaviors remain unclear. We used chronic in vivo 2-photon calcium imaging to monitor the responses of neurons in the V1 of tree shrews learning a Go/No-Go fine orientation discrimination task. We observed increases in neural population measures of discriminability for task-relevant stimuli that correlate with performance and depend on a select subset of neurons with preferred orientations that include the rewarded stimulus and nearby orientations biased away from the non-rewarded stimulus. Learning is accompanied by selective enhancement in the response of these neurons to the rewarded stimulus that further increases their ability to discriminate the task stimuli. These changes persist outside of the trained task and predict observed enhancement and impairment in performance of other discriminations, providing evidence for selective and persistent learning-induced plasticity in the V1, with significant consequences for perception.

Not so spontaneous: Multi-dimensional representations of behaviors and context in sensory areas

Sensory areas are spontaneously active in the absence of sensory stimuli. This spontaneous activity has long been studied; however, its functional role remains largely unknown. Recent advances in technology, allowing large-scale neural recordings in the awake and behaving animal, have transformed our understanding of spontaneous activity. Studies using these recordings have discovered high-dimensional spontaneous activity patterns, correlation between spontaneous activity and behavior, and dissimilarity between spontaneous and sensory-driven activity patterns. These findings are supported by evidence from developing animals, where a transition toward these characteristics is observed as the circuit matures, as well as by evidence from mature animals across species. These newly revealed characteristics call for the formulation of a new role for spontaneous activity in neural sensory computation.

Existing function in primary visual cortex is not perturbed by new skill acquisition of a non-matched sensory task

Acquisition of new skills has the potential to disturb existing network function. To directly assess whether previously acquired cortical function is altered during learning, mice were trained in an abstract task in which selected activity patterns were rewarded using an optical brain-computer interface device coupled to primary visual cortex (V1) neurons. Excitatory neurons were longitudinally recorded using 2-photon calcium imaging. Despite significant changes in local neural activity during task performance, tuning properties and stimulus encoding assessed outside of the trained context were not perturbed. Similarly, stimulus tuning was stable in neurons that remained responsive following a different, visual discrimination training task. However, visual discrimination training increased the rate of representational drift. Our results indicate that while some forms of perceptual learning may modify the contribution of individual neurons to stimulus encoding, new skill learning is not inherently disruptive to the quality of stimulus representation in adult V1.

Norepinephrine potentiates and serotonin depresses visual cortical responses by transforming eligibility traces

Reinforcement allows organisms to learn which stimuli predict subsequent biological relevance. Hebbian mechanisms of synaptic plasticity are insufficient to account for reinforced learning because neuromodulators signaling biological relevance are delayed with respect to the neural activity associated with the stimulus. A theoretical solution is the concept of eligibility traces (eTraces), silent synaptic processes elicited by activity which upon arrival of a neuromodulator are converted into a lasting change in synaptic strength. Previously we demonstrated in visual cortical slices the Hebbian induction of eTraces and their conversion into LTP and LTD by the retroactive action of norepinephrine and serotonin Here we show in vivo in mouse V1 that the induction of eTraces and their conversion to LTP/D by norepinephrine and serotonin respectively potentiates and depresses visual responses. We also show that the integrity of this process is crucial for ocular dominance plasticity, a canonical model of experience-dependent plasticity.

Dynamic Distortion of Orientation Representation after Learning in the Mouse Primary Visual Cortex

Learning is an essential cognitive mechanism allowing behavioral adaptation through adjustments in neuronal processing. It is associated with changes in the activity of sensory cortical neurons evoked by task-relevant stimuli. However, the exact nature of those modifications and the computational advantages they may confer are still debated. Here, we investigated how learning an orientation discrimination task alters the neuronal representations of the cues orientations in the primary visual cortex (V1) of male and female mice. When comparing the activity evoked by the task stimuli in naive mice and the mice performing the task, we found that the representations of the orientation of the rewarded and nonrewarded cues were more accurate and stable in trained mice. This better cue representation in trained mice was associated with a distortion of the orientation representation space such that stimuli flanking the task-relevant orientations were represented as the task stimuli themselves, suggesting that those stimuli were generalized as the task cues. This distortion was context dependent as it was absent in trained mice passively viewing the task cues and enhanced in the behavioral sessions where mice performed best. Those modifications of the V1 population orientation representation in performing mice were supported by a suppression of the activity of neurons tuned for orientations neighboring the orientations of the task cues. Thus, visual processing in V1 is dynamically adapted to enhance the reliability of the representation of the learned cues and favor generalization in the task-relevant computational space.SIGNIFICANCE STATEMENT Performance improvement in a task often requires facilitating the extraction of the information necessary to its execution. Here, we demonstrate the existence of a suppression mechanism that improves the representation of the orientations of the task stimuli in the V1 of mice performing an orientation discrimination task. We also show that this mechanism distorts the V1 orientation representation space, leading stimuli flanking the task stimuli orientations to be generalized as the task stimuli themselves.

Dynamic modulation of neural feedback processing and attention during spatial probabilistic learning

Learned stimulus-reward associations can modulate behavior and the underlying neural processing of information. We investigated the cascade of these neurocognitive mechanisms involved in the learning of spatial stimulus-reward associations. Using electroencephalogram recordings while participants performed a probabilistic spatial reward learning task, we observed that the feedback-related negativity component was more negative in response to loss feedback compared to gain feedback but showed no modulation by learning. The late positive component became larger in response to losses as the learning set progressed but smaller in response to gains. In addition, feedback-locked alpha frequency oscillations measured over occipital sites were predictive of N2pc amplitudes—a marker of spatial attention orienting—observed on the next trial. This relationship was found to become stronger with learning set progression. Taken together, we elucidated neurocognitive dynamics underlying feedback processing during spatial reward learning, and the subsequent effects of these learned spatial stimulus-reward associations on spatial attention.

•

We can learn which spatial location relates to the highest probability of reward

•

Neural processing of feedback valence was not influenced by learning

•

LPC amplitude was dynamically modulated by learning, reflecting context updating

•

Feedback-locked alpha power was predictive of ensuing orientation of attention

An increase in spontaneous activity mediates visual habituation

The cerebral cortex is spontaneously active, but the function of this ongoing activity remains unclear. To test whether spontaneous activity encodes learned experiences, we measured the response of neuronal populations in mouse primary visual cortex with chronic two-photon calcium imaging during visual habituation to a specific oriented stimulus. We find that, during habituation, spontaneous activity increases in neurons across the full range of orientation selectivity, eventually matching that of evoked levels. This increase in spontaneous activity robustly correlates with the degree of habituation. Moreover, boosting spontaneous activity with two-photon optogenetic stimulation to the levels of visually evoked activity accelerates habituation. Our study shows that cortical spontaneous activity is linked to habituation, and we propose that habituation unfolds by minimizing the difference between spontaneous and stimulus-evoked activity levels. We conclude that baseline spontaneous activity could gate incoming sensory information to the cortex based on the learned experience of the animal.

Stimulus-Selective Response Plasticity in Primary Visual Cortex: Progress and Puzzles

Stimulus-selective response plasticity (SRP) is a robust and lasting modification of primary visual cortex (V1) that occurs in response to exposure to novel visual stimuli. It is readily observed as a pronounced increase in the magnitude of visual evoked potentials (VEPs) recorded in response to phase-reversing grating stimuli in neocortical layer 4. The expression of SRP at the individual neuron level is equally robust, but the qualities vary depending on the neuronal type and how activity is measured. This form of plasticity is highly selective for stimulus features such as stimulus orientation, spatial frequency, and contrast. Several key insights into the significance and underlying mechanisms of SRP have recently been made. First, it occurs concomitantly and shares core mechanisms with behavioral habituation, indicating that SRP reflects the formation of long-term familiarity that can support recognition of innocuous stimuli. Second, SRP does not manifest within a recording session but only emerges after an off-line period of several hours that includes sleep. Third, SRP requires not only canonical molecular mechanisms of Hebbian synaptic plasticity within V1, but also the opposing engagement of two key subclasses of cortical inhibitory neuron: the parvalbumin- and somatostatin-expressing GABAergic interneurons. Fourth, pronounced shifts in the power of cortical oscillations from high frequency (gamma) to low frequency (alpha/beta) oscillations provide respective readouts of the engagement of these inhibitory neuronal subtypes following familiarization. In this article we will discuss the implications of these findings and the outstanding questions that remain to gain a deeper understanding of this striking form of experience-dependent plasticity.

Neocortex saves energy by reducing coding precision during food scarcity

Information processing is energetically expensive. In the mammalian brain, it is unclear how information coding and energy use are regulated during food scarcity. Using whole-cell recordings and two-photon imaging in layer 2/3 mouse visual cortex, we found that food restriction reduced AMPA receptor conductance, reducing synaptic ATP use by 29%. Neuronal excitability was nonetheless preserved by a compensatory increase in input resistance and a depolarized resting potential. Consequently, neurons spiked at similar rates as controls but spent less ATP on underlying excitatory currents. This energy-saving strategy had a cost because it amplified the variability of visually-evoked subthreshold responses, leading to a 32% broadening of orientation tuning and impaired fine visual discrimination. This reduction in coding precision was associated with reduced levels of the fat mass-regulated hormone leptin and was restored by exogenous leptin supplementation. Our findings reveal that metabolic state dynamically regulates the energy spent on coding precision in neocortex.

Parametric Copula-GP model for analyzing multidimensional neuronal and behavioral relationships

One of the main goals of current systems neuroscience is to understand how neuronal populations integrate sensory information to inform behavior. However, estimating stimulus or behavioral information that is encoded in high-dimensional neuronal populations is challenging. We propose a method based on parametric copulas which allows modeling joint distributions of neuronal and behavioral variables characterized by different statistics and timescales. To account for temporal or spatial changes in dependencies between variables, we model varying copula parameters by means of Gaussian Processes (GP). We validate the resulting Copula-GP framework on synthetic data and on neuronal and behavioral recordings obtained in awake mice. We show that the use of a parametric description of the high-dimensional dependence structure in our method provides better accuracy in mutual information estimation in higher dimensions compared to other non-parametric methods. Moreover, by quantifying the redundancy between neuronal and behavioral variables, our model exposed the location of the reward zone in an unsupervised manner (i.e., without using any explicit cues about the task structure). These results demonstrate that the Copula-GP framework is particularly useful for the analysis of complex multidimensional relationships between neuronal, sensory and behavioral variables.

Understanding the relationship between a set of variables is a common problem in many fields, such as weather forecast or stock market data. In neuroscience, one of the main challenges is to characterize the dependencies between neuronal activity, sensory stimuli and behavioral outputs. A method of choice for modeling such statistical dependencies is based on copulas, which disentangle dependencies from single variable statistics. To account for changes in dependencies, we model changes in copula parameters by means of Gaussian Processes, conditioned on a task-related variable. The novelty of our approach includes 1) explicit modeling of the dependencies; and 2) combining different copulas to describe experimentally observed variability. We validate the goodness-of-fit as well as information estimates on synthetic data and on recordings from the visual cortex of mice performing a behavioral task. Our parametric model demonstrates significantly better performance in describing high dimensional dependencies compared to other commonly used techniques. We demonstrate that our model can estimate information and predict behaviorally-relevant parameters of the task without providing any explicit cues to the model. Our results indicate that our model is interpretable in the context of neuroscience applications, scalable to large datasets and suitable for accurate statistical modeling and information estimation.

A cortical circuit for audio-visual predictions

Learned associations between stimuli in different sensory modalities can shape the way we perceive these stimuli. However, it is not well understood how these interactions are mediated or at what level of the processing hierarchy they occur. Here we describe a neural mechanism by which an auditory input can shape visual representations of behaviorally relevant stimuli through direct interactions between auditory and visual cortices in mice. We show that the association of an auditory stimulus with a visual stimulus in a behaviorally relevant context leads to experience-dependent suppression of visual responses in primary visual cortex (V1). Auditory cortex axons carry a mixture of auditory and retinotopically matched visual input to V1, and optogenetic stimulation of these axons selectively suppresses V1 neurons that are responsive to the associated visual stimulus after, but not before, learning. Our results suggest that cross-modal associations can be communicated by long-range cortical connections and that, with learning, these cross-modal connections function to suppress responses to predictable input.

FosGFP expression does not capture a sensory learning-related engram in superficial layers of mouse barrel cortex

Immediate-early gene (IEG) expression has been used to identify small neural ensembles linked to a particular experience, based on the principle that a selective subset of activated neurons will encode specific memories or behavioral responses. The majority of these studies have focused on "engrams" in higher-order brain areas where more abstract or convergent sensory information is represented, such as the hippocampus, prefrontal cortex, or amygdala. In primary sensory cortex, IEG expression can label neurons that are responsive to specific sensory stimuli, but experience-dependent shaping of neural ensembles marked by IEG expression has not been demonstrated. Here, we use a fosGFP transgenic mouse to longitudinally monitor in vivo expression of the activity-dependent gene c-fos in superficial layers (L2/3) of primary somatosensory cortex (S1) during a whisker-dependent learning task. We find that sensory association training does not detectably alter fosGFP expression in L2/3 neurons. Although training broadly enhances thalamocortical synaptic strength in pyramidal neurons, we find that synapses onto fosGFP+ neurons are not selectively increased by training; rather, synaptic strengthening is concentrated in fosGFP- neurons. Taken together, these data indicate that expression of the IEG reporter fosGFP does not facilitate identification of a learning-specific engram in L2/3 in barrel cortex during whisker-dependent sensory association learning.

Generalized attention-weighted reinforcement learning

In neuroscience, attention has been shown to bidirectionally interact with reinforcement learning (RL) to reduce the dimensionality of task representations, restricting computations to relevant features. In machine learning, despite their popularity, attention mechanisms have seldom been administered to decision-making problems. Here, we leverage a theoretical model from computational neuroscience - the attention-weighted RL (AWRL), defining how humans identify task-relevant features (i.e., that allow value predictions) - to design an applied deep RL paradigm. We formally demonstrate that the conjunction of the self-attention mechanism, widely employed in machine learning, with value function approximation is a general formulation of the AWRL model. To evaluate our agent, we train it on three Atari tasks at different complexity levels, incorporating both task-relevant and irrelevant features. Because the model uses semantic observations, we can uncover not only which features the agent elects to base decisions on, but also how it chooses to compile more complex, relational features from simpler ones. We first show that performance depends in large part on the ability to compile new compound features, rather than mere focus on individual features. In line with neuroscience predictions, self-attention leads to high resiliency to noise (irrelevant features) compared to other benchmark models. Finally, we highlight the importance and separate contributions of both bottom-up and top-down attention in the learning process. Together, these results demonstrate the broader validity of the AWRL framework in complex task scenarios, and illustrate the benefits of a deeper integration between neuroscience-derived models and RL for decision making in machine learning.

Context value updating and multidimensional neuronal encoding in the retrosplenial cortex

The retrosplenial cortex (RSC) has diverse functional inputs and is engaged by various sensory, spatial, and associative learning tasks. We examine how multiple functional aspects are integrated on the single-cell level in the RSC and how the encoding of task-related parameters changes across learning. Using a visuospatial context discrimination paradigm and two-photon calcium imaging in behaving mice, a large proportion of dysgranular RSC neurons was found to encode multiple task-related dimensions while forming context-value associations across learning. During reversal learning requiring increased cognitive flexibility, we revealed an increased proportion of multidimensional encoding neurons that showed higher decoding accuracy for behaviorally relevant context-value associations. Chemogenetic inactivation of RSC led to decreased behavioral context discrimination during learning phases in which context-value associations were formed, while recall of previously formed associations remained intact. RSC inactivation resulted in a persistent positive behavioral bias in valuing contexts, indicating a role for the RSC in context-value updating.