Differential corticostriatal plasticity during fast and slow motor skill learning in mice

BACKGROUND: Motor skill learning usually comprises "fast" improvement in performance within the initial training session and "slow" improvement that develops across sessions. Previous studies have revealed changes in activity and connectivity in motor cortex and striatum during motor skill learning. However, the nature and dynamics of the plastic changes in each of these brain structures during the different phases of motor learning remain unclear. RESULTS: By using multielectrode arrays, we recorded the simultaneous activity of neuronal ensembles in motor cortex and dorsal striatum of mice during the different phases of skill learning on an accelerating rotarod. Mice exhibited fast improvement in the task during the initial session and also slow improvement across days. Throughout training, a high percentage of striatal (57%) and motor cortex (55%) neurons were task related; i.e., changed their firing rate while mice were running on the rotarod. Improvement in performance was accompanied by substantial plastic changes in both striatum and motor cortex. We observed parallel recruitment of task-related neurons in both structures specifically during the first session. Conversely, during slow learning across sessions we observed differential refinement of the firing patterns in each structure. At the neuronal ensemble level, we observed considerable changes in activity within the first session that became less evident during subsequent sessions. CONCLUSIONS: These data indicate that cortical and striatal circuits exhibit remarkable but dissociable plasticity during fast and slow motor skill learning and suggest that distinct neural processes mediate the different phases of motor skill learning.     

Perceptual learning reduces interneuronal correlations in macaque visual cortex

Responses of neurons in early visual cortex change little with training and appear insufficient to account for perceptual learning. Behavioral performance, however, relies on population activity, and the accuracy of a population code is constrained by correlated noise among neurons. We tested whether training changes interneuronal correlations in the dorsal medial superior temporal area, which is involved in multisensory heading perception. Pairs of single units were recorded simultaneously in two groups of subjects: animals trained extensively in a heading discrimination task, and "naive" animals that performed a passive fixation task. Correlated noise was significantly weaker in trained versus naive animals, which might be expected to improve coding efficiency. However, we show that the observed uniform reduction in noise correlations leads to little change in population coding efficiency when all neurons are decoded. Thus, global changes in correlated noise among sensory neurons may be insufficient to account for perceptual learning.     

Recording Single Neurons' Action Potentials from Freely Moving Pigeons Across Three Stages of Learning

While the subject of learning has attracted immense interest from both behavioral and neural scientists, only relatively few investigators have observed single-neuron activity while animals are acquiring an operantly conditioned response, or when that response is extinguished. But even in these cases, observation periods usually encompass only a single stage of learning, i.e. acquisition or extinction, but not both (exceptions include protocols employing reversal learning; see Bingman et al.¹ for an example). However, acquisition and extinction entail different learning mechanisms and are therefore expected to be accompanied by different types and/or loci of neural plasticity.Accordingly, we developed a behavioral paradigm which institutes three stages of learning in a single behavioral session and which is well suited for the simultaneous recording of single neurons'' action potentials. Animals are trained on a single-interval forced choice task which requires mapping each of two possible choice responses to the presentation of different novel visual stimuli (acquisition). After having reached a predefined performance criterion, one of the two choice responses is no longer reinforced (extinction). Following a certain decrement in performance level, correct responses are reinforced again (reacquisition). By using a new set of stimuli in every session, animals can undergo the acquisition-extinction-reacquisition process repeatedly. Because all three stages of learning occur in a single behavioral session, the paradigm is ideal for the simultaneous observation of the spiking output of multiple single neurons. We use pigeons as model systems, but the task can easily be adapted to any other species capable of conditioned discrimination learning.     

Temporal but Not Spatial Variability during Gait Is Reduced after Selective Dorsal Rhizotomy in Children with Cerebral Palsy

Introduction

Variability in task output is a ubiquitous characteristic that results from non-continuous motor neuron firing during muscular force generation. However, variability can also be attributed to errors in control and coordination of the motor neurons themselves in diseases such as cerebral palsy (CP). Selective dorsal rhizotomy (SDR), a neurosurgical approach to sever sensory nerve roots, is thought to decrease redundant or excessive afferent signalling to intramedullary neurons. In addition to its demonstrated ability to reduce muscular spasticity, we hypothesised that SDR is able to decrease variability during gait, the most frequent functional motor activity of daily living.

Methods

Twelve CP children (aged 6.1±1.3yrs), who underwent SDR and performed gait analysis pre- and 12 months postoperatively, were compared to a control group of eleven typically developing (TD) children. Coefficients of variability as well as mean values were analysed for: temporal variables of gait, spatial parameters and velocity.

Results

Gait parameters of cadence (p = 0.006) and foot progression angle at mid-stance (p = 0.041) changed significantly from pre- to post-SDR. The variability of every temporal parameter was significantly reduced after SDR (p = 0.003–0.049), while it remained generally unchanged for the spatial parameters. Only a small change in gait velocity was observed, but variability in cadence was significantly reduced after SDR (p = 0.015). Almost all parameters changed with a tendency towards normal, but differences between TD and CP children remained in all parameters.

Discussion

The results confirm that SDR improves functional gait performance in children with CP. However, almost exclusively, parameters of temporal variability were significantly improved, leading to the conjecture that temporal variability and spatial variability may be governed independently by the motor cortex. As a result, temporal parameters of task performance may be more vulnerable to disruption, but also more responsive to treatment success of interventions such as SDR.     

A Method for High Fidelity Optogenetic Control of Individual Pyramidal Neurons In vivo

Optogenetic methods have emerged as a powerful tool for elucidating neural circuit activity underlying a diverse set of behaviors across a broad range of species. Optogenetic tools of microbial origin consist of light-sensitive membrane proteins that are able to activate (e.g., channelrhodopsin-2, ChR2) or silence (e.g., halorhodopsin, NpHR) neural activity ingenetically-defined cell types over behaviorally-relevant timescales. We first demonstrate a simple approach for adeno-associated virus-mediated delivery of ChR2 and NpHR transgenes to the dorsal subiculum and prelimbic region of the prefrontal cortex in rat. Because ChR2 and NpHR are genetically targetable, we describe the use of this technology to control the electrical activity of specific populations of neurons (i.e., pyramidal neurons) embedded in heterogeneous tissue with high temporal precision. We describe herein the hardware, custom software user interface, and procedures that allow for simultaneous light delivery and electrical recording from transduced pyramidal neurons in an anesthetized in vivo preparation. These light-responsive tools provide the opportunity for identifying the causal contributions of different cell types to information processing and behavior.     

Task-induced neural covariability as a signature of approximate Bayesian learning and inference

Perception is often characterized computationally as an inference process in which uncertain or ambiguous sensory inputs are combined with prior expectations. Although behavioral studies have shown that observers can change their prior expectations in the context of a task, robust neural signatures of task-specific priors have been elusive. Here, we analytically derive such signatures under the general assumption that the responses of sensory neurons encode posterior beliefs that combine sensory inputs with task-specific expectations. Specifically, we derive predictions for the task-dependence of correlated neural variability and decision-related signals in sensory neurons. The qualitative aspects of our results are parameter-free and specific to the statistics of each task. The predictions for correlated variability also differ from predictions of classic feedforward models of sensory processing and are therefore a strong test of theories of hierarchical Bayesian inference in the brain. Importantly, we find that Bayesian learning predicts an increase in so-called “differential correlations” as the observer’s internal model learns the stimulus distribution, and the observer’s behavioral performance improves. This stands in contrast to classic feedforward encoding/decoding models of sensory processing, since such correlations are fundamentally information-limiting. We find support for our predictions in data from existing neurophysiological studies across a variety of tasks and brain areas. Finally, we show in simulation how measurements of sensory neural responses can reveal information about a subject’s internal beliefs about the task. Taken together, our results reinterpret task-dependent sources of neural covariability as signatures of Bayesian inference and provide new insights into their cause and their function.     

Dynamic patterns of correlated activity in the prefrontal cortex encode information about social behavior

New technologies make it possible to measure activity from many neurons simultaneously. One approach is to analyze simultaneously recorded neurons individually, then group together neurons which increase their activity during similar behaviors into an “ensemble.” However, this notion of an ensemble ignores the ability of neurons to act collectively and encode and transmit information in ways that are not reflected by their individual activity levels. We used microendoscopic GCaMP imaging to measure prefrontal activity while mice were either alone or engaged in social interaction. We developed an approach that combines a neural network classifier and surrogate (shuffled) datasets to characterize how neurons synergistically transmit information about social behavior. Notably, unlike optimal linear classifiers, a neural network classifier with a single linear hidden layer can discriminate network states which differ solely in patterns of coactivity, and not in the activity levels of individual neurons. Using this approach, we found that surrogate datasets which preserve behaviorally specific patterns of coactivity (correlations) outperform those which preserve behaviorally driven changes in activity levels but not correlated activity. Thus, social behavior elicits increases in correlated activity that are not explained simply by the activity levels of the underlying neurons, and prefrontal neurons act collectively to transmit information about socialization via these correlations. Notably, this ability of correlated activity to enhance the information transmitted by neuronal ensembles is diminished in mice lacking the autism-associated gene Shank3. These results show that synergy is an important concept for the coding of social behavior which can be disrupted in disease states, reveal a specific mechanism underlying this synergy (social behavior increases correlated activity within specific ensembles), and outline methods for studying how neurons within an ensemble can work together to encode information.

Behaviorally-specific patterns of correlated activity between prefrontal neurons normally enhance the information that neuronal ensembles transmit about social behavior. This study shows that in a mouse model of autism, individual neurons continue to encode social information, but this additional information carried by patterns of correlated activity is lost.     

The neurosteroid allopregnanolone impairs object memory and contextual fear memory in male C57BL/6J mice

Allopregnanolone (ALLO, or 3α-hydroxy-5α-pregnan-20-one) is a steroid metabolite of progesterone and a potent endogenous positive allosteric modulator of GABA-A receptors. Systemic ALLO has been reported to impair spatial, but not nonspatial learning in the Morris water maze (MWM) and contextual memory in rodents. These cognitive effects suggest an influence of ALLO on hippocampal-dependent memory, although the specific nature of the neurosteroid's effects on learning, memory or performance is unclear. The present studies aimed to determine: (i) the memory process(es) affected by systemic ALLO using a nonspatial object memory task; and (ii) whether ALLO affects object memory via an influence within the dorsal hippocampus. Male C57BL/6J mice received systemic ALLO either before or immediately after the sample session of a novel object recognition (NOR) task. Results demonstrated that systemic ALLO impaired the encoding and consolidation of object memory. A subsequent study revealed that bilateral microinfusion of ALLO into the CA1 region of dorsal hippocampus immediately following the NOR sample session also impaired object memory consolidation. In light of debate over the hippocampal-dependence of object recognition memory, we also tested systemic ALLO-treated mice on a contextual and cued fear-conditioning task. Systemic ALLO impaired the encoding of contextual memory when administered prior to the context pre-exposure session. Together, these results indicate that ALLO exhibits primary effects on memory encoding and consolidation, and extend previous findings by demonstrating a sensitivity of nonspatial memory to ALLO, likely by disrupting dorsal hippocampal function.     

Repeated Stimulus Exposure Alters the Way Sound Is Encoded in the Human Brain

Auditory training programs are being developed to remediate various types of communication disorders. Biological changes have been shown to coincide with improved perception following auditory training so there is interest in determining if these changes represent biologic markers of auditory learning. Here we examine the role of stimulus exposure and listening tasks, in the absence of training, on the modulation of evoked brain activity. Twenty adults were divided into two groups and exposed to two similar sounding speech syllables during four electrophysiological recording sessions (24 hours, one week, and up to one year later). In between each session, members of one group were asked to identify each stimulus. Both groups showed enhanced neural activity from session-to-session, in the same P2 latency range previously identified as being responsive to auditory training. The enhancement effect was most pronounced over temporal-occipital scalp regions and largest for the group who participated in the identification task. The effects were rapid and long-lasting with enhanced synchronous activity persisting months after the last auditory experience. Physiological changes did not coincide with perceptual changes so results are interpreted to mean stimulus exposure, with and without being paired with an identification task, alters the way sound is processed in the brain. The cumulative effect likely involves auditory memory; however, in the absence of training, the observed physiological changes are insufficient to result in changes in learned behavior.     

On the Number of Neurons and Time Scale of Integration Underlying the Formation of Percepts in the Brain

All of our perceptual experiences arise from the activity of neural populations. Here we study the formation of such percepts under the assumption that they emerge from a linear readout, i.e., a weighted sum of the neurons’ firing rates. We show that this assumption constrains the trial-to-trial covariance structure of neural activities and animal behavior. The predicted covariance structure depends on the readout parameters, and in particular on the temporal integration window w and typical number of neurons K used in the formation of the percept. Using these predictions, we show how to infer the readout parameters from joint measurements of a subject’s behavior and neural activities. We consider three such scenarios: (1) recordings from the complete neural population, (2) recordings of neuronal sub-ensembles whose size exceeds K, and (3) recordings of neuronal sub-ensembles that are smaller than K. Using theoretical arguments and artificially generated data, we show that the first two scenarios allow us to recover the typical spatial and temporal scales of the readout. In the third scenario, we show that the readout parameters can only be recovered by making additional assumptions about the structure of the full population activity. Our work provides the first thorough interpretation of (feed-forward) percept formation from a population of sensory neurons. We discuss applications to experimental recordings in classic sensory decision-making tasks, which will hopefully provide new insights into the nature of perceptual integration.     

Detection of neural activity in the brains of Japanese honeybee workers during the formation of a "hot defensive bee ball"

Anti-predator behaviors are essential to survival for most animals. The neural bases of such behaviors, however, remain largely unknown. Although honeybees commonly use their stingers to counterattack predators, the Japanese honeybee (Apis cerana japonica) uses a different strategy to fight against the giant hornet (Vespa mandarinia japonica). Instead of stinging the hornet, Japanese honeybees form a “hot defensive bee ball” by surrounding the hornet en masse, killing it with heat. The European honeybee (A. mellifera ligustica), on the other hand, does not exhibit this behavior, and their colonies are often destroyed by a hornet attack. In the present study, we attempted to analyze the neural basis of this behavior by mapping the active brain regions of Japanese honeybee workers during the formation of a hot defensive bee ball. First, we identified an A. cerana homolog (Acks = Apis cerana kakusei) of kakusei, an immediate early gene that we previously identified from A. mellifera, and showed that Acks has characteristics similar to kakusei and can be used to visualize active brain regions in A. cerana. Using Acks as a neural activity marker, we demonstrated that neural activity in the mushroom bodies, especially in Class II Kenyon cells, one subtype of mushroom body intrinsic neurons, and a restricted area between the dorsal lobes and the optic lobes was increased in the brains of Japanese honeybee workers involved in the formation of a hot defensive bee ball. In addition, workers exposed to 46°C heat also exhibited Acks expression patterns similar to those observed in the brains of workers involved in the formation of a hot defensive bee ball, suggesting that the neural activity observed in the brains of workers involved in the hot defensive bee ball mainly reflects thermal stimuli processing.     

Democratic population decisions result in robust policy-gradient learning: a parametric study with GPU simulations

High performance computing on the Graphics Processing Unit (GPU) is an emerging field driven by the promise of high computational power at a low cost. However, GPU programming is a non-trivial task and moreover architectural limitations raise the question of whether investing effort in this direction may be worthwhile. In this work, we use GPU programming to simulate a two-layer network of Integrate-and-Fire neurons with varying degrees of recurrent connectivity and investigate its ability to learn a simplified navigation task using a policy-gradient learning rule stemming from Reinforcement Learning. The purpose of this paper is twofold. First, we want to support the use of GPUs in the field of Computational Neuroscience. Second, using GPU computing power, we investigate the conditions under which the said architecture and learning rule demonstrate best performance. Our work indicates that networks featuring strong Mexican-Hat-shaped recurrent connections in the top layer, where decision making is governed by the formation of a stable activity bump in the neural population (a "non-democratic" mechanism), achieve mediocre learning results at best. In absence of recurrent connections, where all neurons "vote" independently ("democratic") for a decision via population vector readout, the task is generally learned better and more robustly. Our study would have been extremely difficult on a desktop computer without the use of GPU programming. We present the routines developed for this purpose and show that a speed improvement of 5x up to 42x is provided versus optimised Python code. The higher speed is achieved when we exploit the parallelism of the GPU in the search of learning parameters. This suggests that efficient GPU programming can significantly reduce the time needed for simulating networks of spiking neurons, particularly when multiple parameter configurations are investigated.     

Dynamics of population code for working memory in the prefrontal cortex

Some neurons (delay cells) in the prefrontal cortex elevate their activities throughout the time period during which the animal is required to remember past events and prepare future behavior, suggesting that working memory is mediated by continuous neural activity. It is unknown, however, how working memory is represented within a population of prefrontal cortical neurons. We recorded from neuronal ensembles in the prefrontal cortex as rats learned a new delayed alternation task. Ensemble activities changed in parallel with behavioral learning so that they increasingly allowed correct decoding of previous and future goal choices. In well-trained rats, considerable decoding was possible based on only a few neurons and after removing continuously active delay cells. These results show that neural activity in the prefrontal cortex changes dynamically during new task learning so that working memory is robustly represented and that working memory can be mediated by sequential activation of different neural populations.     

Learning neural connectivity from firing activity: efficient algorithms with provable guarantees on topology

The connectivity of a neuronal network has a major effect on its functionality and role. It is generally believed that the complex network structure of the brain provides a physiological basis for information processing. Therefore, identifying the network’s topology has received a lot of attentions in neuroscience and has been the center of many research initiatives such as Human Connectome Project. Nevertheless, direct and invasive approaches that slice and observe the neural tissue have proven to be time consuming, complex and costly. As a result, the inverse methods that utilize firing activity of neurons in order to identify the (functional) connections have gained momentum recently, especially in light of rapid advances in recording technologies; It will soon be possible to simultaneously monitor the activities of tens of thousands of neurons in real time. While there are a number of excellent approaches that aim to identify the functional connections from firing activities, the scalability of the proposed techniques plays a major challenge in applying them on large-scale datasets of recorded firing activities. In exceptional cases where scalability has not been an issue, the theoretical performance guarantees are usually limited to a specific family of neurons or the type of firing activities. In this paper, we formulate the neural network reconstruction as an instance of a graph learning problem, where we observe the behavior of nodes/neurons (i.e., firing activities) and aim to find the links/connections. We develop a scalable learning mechanism and derive the conditions under which the estimated graph for a network of Leaky Integrate and Fire (LIf) neurons matches the true underlying synaptic connections. We then validate the performance of the algorithm using artificially generated data (for benchmarking) and real data recorded from multiple hippocampal areas in rats.     

Tool Use Specific Adult Neurogenesis and Synaptogenesis in Rodent (Octodon degus) Hippocampus

We previously demonstrated that degus (Octodon degus), which are a species of small caviomorph rodents, could be trained to use a T-shaped rake as a hand tool to expand accessible spaces. To elucidate the neurobiological underpinnings of this higher brain function, we compared this tool use learning task with a simple spatial (radial maze) memory task and investigated the changes that were induced in the hippocampal neural circuits known to subserve spatial perception and learning. With the exposure to an enriched environment in home cage, adult neurogenesis in the dentate gyrus of the hippocampus was augmented by tool use learning, but not radial maze learning, when compared to control conditions. Furthermore, the proportion of new synapses formed in the CA3 region of the hippocampus, the target area for projections of mossy fiber axons emanating from newborn neurons, was specifically increased by tool use learning. Thus, active tool use behavior by rodents, learned through multiple training sessions, requires the hippocampus to generate more novel neurons and synapses than spatial information processing in radial maze learning.     

Aplysia Ganglia Preparation for Electrophysiological and Molecular Analyses of Single Neurons

A major challenge in neurobiology is to understand the molecular underpinnings of neural circuitry that govern a specific behavior. Once the specific molecular mechanisms are identified, new therapeutic strategies can be developed to treat abnormalities in specific behaviors caused by degenerative diseases or aging of the nervous system. The marine snail Aplysia californica is well suited for the investigations of cellular and molecular basis of behavior because neural circuitry underlying a specific behavior could be easily determined and the individual components of the circuitry could be easily manipulated. These advantages of Aplysia have led to several fundamental discoveries of neurobiology of learning and memory. Here we describe a preparation of the Aplysia nervous system for the electrophysiological and molecular analyses of individual neurons. Briefly, ganglion dissected from the nervous system is exposed to protease to remove the ganglion sheath such that neurons are exposed but retain neuronal activity as in the intact animal. This preparation is used to carry out electrophysiological measurements of single or multiple neurons. Importantly, following the recording using a simple methodology, the neurons could be isolated directly from the ganglia for gene expression analysis. These protocols were used to carry out simultaneous electrophysiological recordings from L7 and R15 neurons, study their response to acetylcholine and quantitating expression of CREB1 gene in isolated single L7, L11, R15, and R2 neurons of Aplysia.     

Neural mechanisms of human perceptual learning: electrophysiological evidence for a two-stage process

Background

Humans and other animals change the way they perceive the world due to experience. This process has been labeled as perceptual learning, and implies that adult nervous systems can adaptively modify the way in which they process sensory stimulation. However, the mechanisms by which the brain modifies this capacity have not been sufficiently analyzed.

Methodology/Principal Findings

We studied the neural mechanisms of human perceptual learning by combining electroencephalographic (EEG) recordings of brain activity and the assessment of psychophysical performance during training in a visual search task. All participants improved their perceptual performance as reflected by an increase in sensitivity (d'') and a decrease in reaction time. The EEG signal was acquired throughout the entire experiment revealing amplitude increments, specific and unspecific to the trained stimulus, in event-related potential (ERP) components N2pc and P3 respectively. P3 unspecific modification can be related to context or task-based learning, while N2pc may be reflecting a more specific attentional-related boosting of target detection. Moreover, bell and U-shaped profiles of oscillatory brain activity in gamma (30–60 Hz) and alpha (8–14 Hz) frequency bands may suggest the existence of two phases for learning acquisition, which can be understood as distinctive optimization mechanisms in stimulus processing.

Conclusions/Significance

We conclude that there are reorganizations in several neural processes that contribute differently to perceptual learning in a visual search task. We propose an integrative model of neural activity reorganization, whereby perceptual learning takes place as a two-stage phenomenon including perceptual, attentional and contextual processes.     

Cue Reactivity Is Associated with Duration and Severity of Alcohol Dependence: An fMRI Study

Introduction

With the progression of substance dependence, drug cue-related brain activation is thought to shift from motivational towards habit pathways. However, a direct association between cue-induced brain activation and dependence duration has not yet been shown. We therefore examined the relationship between alcohol cue-reactivity in the brain, cue-induced subjective craving and alcohol dependence duration and severity. Since alcohol dependence is highly comorbid with depression/anxiety, which may modulate brain responses to alcohol cues, we also examined the relation between comorbid depression/anxiety and cue-reactivity.

Methods

We compared 30 alcohol dependent patients with 15 healthy controls and 15 depression/anxiety patients during a visual alcohol cue-reactivity task using functional magnetic resonance imaging blood oxygenated level-dependent responses and subjective craving as outcomes. Within the alcohol dependent group we correlated cue-reactivity with alcohol dependence severity and duration, with cue-induced craving and with depression/anxiety levels.

Results

Alcohol dependent patients showed greater cue-reactivity in motivational brain pathways and stronger subjective craving than depression/anxiety patients and healthy controls. Depression/anxiety was not associated with cue-reactivity, but depression severity in alcohol dependent patients was positively associated with craving. Within alcohol dependence, longer duration of alcohol dependence was associated with stronger cue-related activation of the posterior putamen, a structure involved in habits, whereas higher alcohol dependence severity was associated with lower cue-reactivity in the anterior putamen, an area implicated in goal-directed behavior preceding habit formation.

Conclusion

Cue-reactivity in alcohol dependence is not modulated by comorbid depression or anxiety. More importantly, the current data confirm the hypothesis of a ventral to dorsal striatal shift of learning processes with longer dependence duration, which could underlie increasingly habitual substance use with progressing substance dependence.     

Neuromolecular correlates of cooperation and conflict during territory defense in a cichlid fish

Cooperative behavior is widespread among animals, yet the neural mechanisms have not been studied in detail. We examined cooperative territory defense behavior and associated neural activity in candidate forebrain regions in the cichlid fish, Astatotilapia burtoni. We find that a territorial male neighbor will engage in territory defense dependent on the perceived threat of the intruder. The resident male, on the other hand, engages in defense based on the size and behavior of his partner, the neighbor. In the neighbor, we find that an index of engagement correlates with neural activity in the putative homolog of the mammalian basolateral amygdala and in the preoptic area, as well as in preoptic dopaminergic neurons. In the resident, neighbor behavior is correlated with neural activity in the homolog of the mammalian hippocampus. Overall, we find distinct neural activity patterns between the neighbor and the resident, suggesting that an individual perceives and processes an intruder challenge differently during cooperative territory defense depending on its own behavioral role.