Emergence of Coordinated Neural Dynamics Underlies Neuroprosthetic Learning and Skillful Control

Selective consistency of recurrent neural networks induced by plasticity as a mechanism of unsupervised perceptual learning

Understanding the mechanism by which the brain achieves relatively consistent information processing contrary to its inherent inconsistency in activity is one of the major challenges in neuroscience. Recently, it has been reported that the consistency of neural responses to stimuli that are presented repeatedly is enhanced implicitly in an unsupervised way, and results in improved perceptual consistency. Here, we propose the term "selective consistency" to describe this input-dependent consistency and hypothesize that it will be acquired in a self-organizing manner by plasticity within the neural system. To test this, we investigated whether a reservoir-based plastic model could acquire selective consistency to repeated stimuli. We used white noise sequences randomly generated in each trial and referenced white noise sequences presented multiple times. The results showed that the plastic network was capable of acquiring selective consistency rapidly, with as little as five exposures to stimuli, even for white noise. The acquisition of selective consistency could occur independently of performance optimization, as the network's time-series prediction accuracy for referenced stimuli did not improve with repeated exposure and optimization. Furthermore, the network could only achieve selective consistency when in the region between order and chaos. These findings suggest that the neural system can acquire selective consistency in a self-organizing manner and that this may serve as a mechanism for certain types of learning.

Transitioning from global to local computational strategies during brain-machine interface learning

When learning to use a brain-machine interface (BMI), the brain modulates neuronal activity patterns, exploring and exploiting the state space defined by their neural manifold. Neurons directly involved in BMI control (i.e., direct neurons) can display marked changes in their firing patterns during BMI learning. However, the extent of firing pattern changes in neurons not directly involved in BMI control (i.e., indirect neurons) remains unclear. To clarify this issue, we localized direct and indirect neurons to separate hemispheres in a task designed to bilaterally engage these hemispheres while animals learned to control the position of a platform with their neural signals. Animals that learned to control the platform and improve their performance in the task shifted from a global strategy, where both direct and indirect neurons modified their firing patterns, to a local strategy, where only direct neurons modified their firing rate, as animals became expert in the task. Animals that did not learn the BMI task did not shift from utilizing a global to a local strategy. These results provide important insights into what differentiates successful and unsuccessful BMI learning and the computational mechanisms adopted by the neurons.

Cortico-cerebellar coordination facilitates neuroprosthetic control

Temporally coordinated neural activity is central to nervous system function and purposeful behavior. Still, there is a paucity of evidence demonstrating how this coordinated activity within cortical and subcortical regions governs behavior. We investigated this between the primary motor (M1) and contralateral cerebellar cortex as rats learned a neuroprosthetic/brain-machine interface (BMI) task. In neuroprosthetic task, actuator movements are causally linked to M1 "direct" neurons that drive the decoder for successful task execution. However, it is unknown how task-related M1 activity interacts with the cerebellum. We observed a notable 3 to 6 hertz coherence that emerged between these regions' local field potentials (LFPs) with learning that also modulated task-related spiking. We identified robust task-related indirect modulation in the cerebellum, which developed a preferential relationship with M1 task-related activity. Inhibiting cerebellar cortical and deep nuclei activity through optogenetics led to performance impairments in M1-driven neuroprosthetic control. Together, these results demonstrate that cerebellar influence is necessary for M1-driven neuroprosthetic control.

Adaptation and learning as strategies to maximize reward in neurofeedback tasks

Introduction

Adaptation and learning have been observed to contribute to the acquisition of new motor skills and are used as strategies to cope with changing environments. However, it is hard to determine the relative contribution of each when executing goal directed motor tasks. This study explores the dynamics of neural activity during a center-out reaching task with continuous visual feedback under the influence of rotational perturbations.

Methods

Results for a brain-computer interface (BCI) task performed by two non-human primate (NHP) subjects are compared to simulations from a reinforcement learning agent performing an analogous task. We characterized baseline activity and compared it to the activity after rotational perturbations of different magnitudes were introduced. We employed principal component analysis (PCA) to analyze the spiking activity driving the cursor in the NHP BCI task as well as the activation of the neural network of the reinforcement learning agent.

Results and discussion

Our analyses reveal that both for the NHPs and the reinforcement learning agent, the task-relevant neural manifold is isomorphic with the task. However, for the NHPs the manifold is largely preserved for all rotational perturbations explored and adaptation of neural activity occurs within this manifold as rotations are compensated by reassignment of regions of the neural space in an angular pattern that cancels said rotations. In contrast, retraining the reinforcement learning agent to reach the targets after rotation results in substantial modifications of the underlying neural manifold. Our findings demonstrate that NHPs adapt their existing neural dynamic repertoire in a quantitatively precise manner to account for perturbations of different magnitudes and they do so in a way that obviates the need for extensive learning.

Assistive sensory-motor perturbations influence learned neural representations

Task errors are used to learn and refine motor skills. We investigated how task assistance influences learned neural representations using Brain-Computer Interfaces (BCIs), which map neural activity into movement via a decoder. We analyzed motor cortex activity as monkeys practiced BCI with a decoder that adapted to improve or maintain performance over days. Population dimensionality remained constant or increased with learning, counter to trends with non-adaptive BCIs. Yet, over time, task information was contained in a smaller subset of neurons or population modes. Moreover, task information was ultimately stored in neural modes that occupied a small fraction of the population variance. An artificial neural network model suggests the adaptive decoders contribute to forming these compact neural representations. Our findings show that assistive decoders manipulate error information used for long-term learning computations, like credit assignment, which informs our understanding of motor learning and has implications for designing real-world BCIs.

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.

Emergence of task-related spatiotemporal population dynamics in transplanted neurons

Loss of nervous system tissue after severe brain injury is a main determinant of poor functional recovery. Cell transplantation is a promising method to restore lost tissue and function, yet it remains unclear if transplanted neurons can demonstrate the population level dynamics important for movement control. Here we present a comprehensive approach for long-term single neuron monitoring and manipulation of transplanted embryonic cortical neurons after cortical injury in adult male mice performing a prehension task. The observed patterns of population activity in the transplanted network strongly resembled that of healthy networks. Specifically, the task-related spatiotemporal activity patterns of transplanted neurons could be represented by latent factors that evolve within a low dimensional manifold. We also demonstrate reliable modulation of the transplanted networks using minimally invasive epidural stimulation. Our approach may allow greater insight into how restoration of cell-type specific network dynamics in vivo can restore motor function.

Neuronal functional connectivity is impaired in a layer dependent manner near the chronically implanted microelectrodes

Objective

This study aims to reveal longitudinal changes in functional network connectivity within and across different brain structures near the chronically implanted microelectrode. While it is well established that the foreign-body response (FBR) contributes to the gradual decline of the signals recorded from brain implants over time, how does the FBR impact affect the functional stability of neural circuits near implanted Brain-Computer Interfaces (BCIs) remains unknown. This research aims to illuminate how the chronic FBR can alter local neural circuit function and the implications for BCI decoders.

Approach

This study utilized multisite Michigan-style microelectrodes that span all cortical layers and the hippocampal CA1 region to collect spontaneous and visually-evoked electrophysiological activity. Alterations in neuronal activity near the microelectrode were tested assessing cross-frequency synchronization of LFP and spike entrainment to LFP oscillatory activity throughout 16 weeks after microelectrode implantation.

Main Results

The study found that cortical layer 4, the input-receiving layer, maintained activity over the implantation time. However, layers 2/3 rapidly experienced severe impairment, leading to a loss of proper intralaminar connectivity in the downstream output layers 5/6. Furthermore, the impairment of interlaminar connectivity near the microelectrode was unidirectional, showing decreased connectivity from Layers 2/3 to Layers 5/6 but not the reverse direction. In the hippocampus, CA1 neurons gradually became unable to properly entrain to the surrounding LFP oscillations.

Significance

This study provides a detailed characterization of network connectivity dysfunction over long-term microelectrode implantation periods. This new knowledge could contribute to the development of targeted therapeutic strategies aimed at improving the health of the tissue surrounding brain implants and potentially inform engineering of adaptive decoders as the FBR progresses. Our study's understanding of the dynamic changes in the functional network over time opens the door to developing interventions for improving the long-term stability and performance of intracortical microelectrodes.

Distinct neural representations during a brain-machine interface and manual reaching task in motor cortex, prefrontal cortex, and striatum

Although brain-machine interfaces (BMIs) are directly controlled by the modulation of a select local population of neurons, distributed networks consisting of cortical and subcortical areas have been implicated in learning and maintaining control. Previous work in rodents has demonstrated the involvement of the striatum in BMI learning. However, the prefrontal cortex has been largely ignored when studying motor BMI control despite its role in action planning, action selection, and learning abstract tasks. Here, we compare local field potentials simultaneously recorded from primary motor cortex (M1), dorsolateral prefrontal cortex (DLPFC), and the caudate nucleus of the striatum (Cd) while nonhuman primates perform a two-dimensional, self-initiated, center-out task under BMI control and manual control. Our results demonstrate the presence of distinct neural representations for BMI and manual control in M1, DLPFC, and Cd. We find that neural activity from DLPFC and M1 best distinguishes control types at the go cue and target acquisition, respectively, while M1 best predicts target-direction at both task events. We also find effective connectivity from DLPFC → M1 throughout both control types and Cd → M1 during BMI control. These results suggest distributed network activity between M1, DLPFC, and Cd during BMI control that is similar yet distinct from manual control.

Brain-machine interface learning is facilitated by specific patterning of distributed cortical feedback

Neuroprosthetics offer great hope for motor-impaired patients. One obstacle is that fine motor control requires near-instantaneous, rich somatosensory feedback. Such distributed feedback may be recreated in a brain-machine interface using distributed artificial stimulation across the cortical surface. Here, we hypothesized that neuronal stimulation must be contiguous in its spatiotemporal dynamics to be efficiently integrated by sensorimotor circuits. Using a closed-loop brain-machine interface, we trained head-fixed mice to control a virtual cursor by modulating the activity of motor cortex neurons. We provided artificial feedback in real time with distributed optogenetic stimulation patterns in the primary somatosensory cortex. Mice developed a specific motor strategy and succeeded to learn the task only when the optogenetic feedback pattern was spatially and temporally contiguous while it moved across the topography of the somatosensory cortex. These results reveal spatiotemporal properties of the sensorimotor cortical integration that set constraints on the design of neuroprosthetics.

Flexible regulation of representations on a drifting manifold enables long-term stable complex neuroprosthetic control

The nervous system needs to balance the stability of neural representations with plasticity. It is unclear what is the representational stability of simple actions, particularly those that are well-rehearsed in humans, and how it changes in new contexts. Using an electrocorticography brain-computer interface (BCI), we found that the mesoscale manifold and relative representational distances for a repertoire of simple imagined movements were remarkably stable. Interestingly, however, the manifold's absolute location demonstrated day-to-day drift. Strikingly, representational statistics, especially variance, could be flexibly regulated to increase discernability during BCI control without somatotopic changes. Discernability strengthened with practice and was specific to the BCI, demonstrating remarkable contextual specificity. Accounting for drift, and leveraging the flexibility of representations, allowed neuroprosthetic control of a robotic arm and hand for over 7 months without recalibration. Our study offers insight into how electrocorticography can both track representational statistics across long periods and allow long-term complex neuroprosthetic control.

Invariant neural dynamics drive commands to control different movements

It has been proposed that the nervous system has the capacity to generate a wide variety of movements because it reuses some invariant code. Previous work has identified that dynamics of neural population activity are similar during different movements, where dynamics refer to how the instantaneous spatial pattern of population activity changes in time. Here, we test whether invariant dynamics of neural populations are actually used to issue the commands that direct movement. Using a brain-machine interface (BMI) that transforms rhesus macaques' motor-cortex activity into commands for a neuroprosthetic cursor, we discovered that the same command is issued with different neural-activity patterns in different movements. However, these different patterns were predictable, as we found that the transitions between activity patterns are governed by the same dynamics across movements. These invariant dynamics are low dimensional, and critically, they align with the BMI, so that they predict the specific component of neural activity that actually issues the next command. We introduce a model of optimal feedback control (OFC) that shows that invariant dynamics can help transform movement feedback into commands, reducing the input that the neural population needs to control movement. Altogether our results demonstrate that invariant dynamics drive commands to control a variety of movements and show how feedback can be integrated with invariant dynamics to issue generalizable commands.

Tracking the Dynamic Neural Connectivity via Conjugate Gradient Optimization. ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY. IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY. ANNUAL INTERNATIONAL CONFERENCE 2023;2023:1-4. [PMID: 38082697 DOI: 10.1109/embc40787.2023.10340664] [Citation(s) in RCA: 0] [Impact Index Per Article: 0]

Neural connectivity describes how neuron populations coordinate and create cognitive and behavioral functions. Neural connectivity performs dynamics where its population spiking responses to stimuli or intention change over time. Brain-machine interface (BMI) provides a framework for studying dynamical neural connectivity. In BMI, point process is a powerful technique in analyzing the single neuronal tuning. And generalized linear mode (GLM) as an encoding model can incorporate the tuning in kinematics and the neural connectivity. Quantification and tracking of dynamic neural connectivity can contribute to the elucidation of the generation of brain functions in a computational way. However, most of the previous work focused on single neuronal adaptation to kinematics. When a neuron is significantly modulated by some other neurons in some tasks, the shape of the log likelihood function for single neuronal observations can be narrowed in some dimensions. And the existing gradient-based methods are not able to reach the optimum in a fast and adaptive searching way. In this work, to maximize the likelihood of observations and obtain the dynamic neural connectivity tuning parameters, we proposed a conjugate gradient-based encoding model (CGE). We illustrate CGE for likelihood function using the real experimental data under manual control and brain control. The results show that the proposed CGE has better performance in tracking the dynamic neural connectivity tuning parameters and modeling neural encoding.Clinical Relevance- Not directly related.

Neural Plasticity in Sensorimotor Brain-Machine Interfaces

Brain-machine interfaces (BMIs) aim to treat sensorimotor neurological disorders by creating artificial motor and/or sensory pathways. Introducing artificial pathways creates new relationships between sensory input and motor output, which the brain must learn to gain dexterous control. This review highlights the role of learning in BMIs to restore movement and sensation, and discusses how BMI design may influence neural plasticity and performance. The close integration of plasticity in sensory and motor function influences the design of both artificial pathways and will be an essential consideration for bidirectional devices that restore both sensory and motor function.

Rapid adaptation of brain-computer interfaces to new neuronal ensembles or participants via generative modelling

For brain-computer interfaces (BCIs), obtaining sufficient training data for algorithms that map neural signals onto actions can be difficult, expensive or even impossible. Here we report the development and use of a generative model-a model that synthesizes a virtually unlimited number of new data distributions from a learned data distribution-that learns mappings between hand kinematics and the associated neural spike trains. The generative spike-train synthesizer is trained on data from one recording session with a monkey performing a reaching task and can be rapidly adapted to new sessions or monkeys by using limited additional neural data. We show that the model can be adapted to synthesize new spike trains, accelerating the training and improving the generalization of BCI decoders. The approach is fully data-driven, and hence, applicable to applications of BCIs beyond motor control.

Long-term unsupervised recalibration of cursor BCIs

Intracortical brain-computer interfaces (iBCIs) require frequent recalibration to maintain robust performance due to changes in neural activity that accumulate over time. Compensating for this nonstationarity would enable consistently high performance without the need for supervised recalibration periods, where users cannot engage in free use of their device. Here we introduce a hidden Markov model (HMM) to infer what targets users are moving toward during iBCI use. We then retrain the system using these inferred targets, enabling unsupervised adaptation to changing neural activity. Our approach outperforms the state of the art in large-scale, closed-loop simulations over two months and in closed-loop with a human iBCI user over one month. Leveraging an offline dataset spanning five years of iBCI recordings, we further show how recently proposed data distribution-matching approaches to recalibration fail over long time scales; only target-inference methods appear capable of enabling long-term unsupervised recalibration. Our results demonstrate how task structure can be used to bootstrap a noisy decoder into a highly-performant one, thereby overcoming one of the major barriers to clinically translating BCIs.

Cortical-hippocampal coupling during manifold exploration in motor cortex

Systems consolidation-a process for long-term memory stabilization-has been hypothesized to occur in two stages1-4. Whereas new memories require the hippocampus5-9, they become integrated into cortical networks over time10-12, making them independent of the hippocampus. How hippocampal-cortical dialogue precisely evolves during this and how cortical representations change in concert is unknown. Here, we use a skill learning task13,14 to monitor the dynamics of cross-area coupling during non-rapid eye movement sleep along with changes in primary motor cortex (M1) representational stability. Our results indicate that precise cross-area coupling between hippocampus, prefrontal cortex and M1 can demarcate two distinct stages of processing. We specifically find that each animal demonstrates a sharp increase in prefrontal cortex and M1 sleep slow oscillation coupling with stabilization of performance. This sharp increase then predicts a drop in hippocampal sharp-wave ripple (SWR)-M1 slow oscillation coupling-suggesting feedback to inform hippocampal disengagement and transition to a second stage. Notably, the first stage shows significant increases in hippocampal SWR-M1 slow oscillation coupling in the post-training sleep and is closely associated with rapid learning and variability of the M1 low-dimensional manifold. Strikingly, even after consolidation, inducing new manifold exploration by changing task parameters re-engages hippocampal-M1 coupling. We thus find evidence for dynamic hippocampal-cortical dialogue associated with manifold exploration during learning and adaptation.

Clinical neuroscience and neurotechnology: An amazing symbiosis

In the last decades, clinical neuroscience found a novel ally in neurotechnologies, devices able to record and stimulate electrical activity in the nervous system. These technologies improved the ability to diagnose and treat neural disorders. Neurotechnologies are concurrently enabling a deeper understanding of healthy and pathological dynamics of the nervous system through stimulation and recordings during brain implants. On the other hand, clinical neurosciences are not only driving neuroengineering toward the most relevant clinical issues, but are also shaping the neurotechnologies thanks to clinical advancements. For instance, understanding the etiology of a disease informs the location of a therapeutic stimulation, but also the way stimulation patterns should be designed to be more effective/naturalistic. Here, we describe cases of fruitful integration such as Deep Brain Stimulation and cortical interfaces to highlight how this symbiosis between clinical neuroscience and neurotechnology is closer to a novel integrated framework than to a simple interdisciplinary interaction.

Selective modulation of cortical population dynamics during neuroprosthetic skill learning

Brain-machine interfaces (BMIs) provide a framework for studying how cortical population dynamics evolve over learning in a task in which the mapping between neural activity and behavior is precisely defined. Learning to control a BMI is associated with the emergence of coordinated neural dynamics in populations of neurons whose activity serves as direct input to the BMI decoder (direct subpopulation). While previous work shows differential modification of firing rate modulation in this population relative to a population whose activity was not directly input to the BMI decoder (indirect subpopulation), little is known about how learning-related changes in cortical population dynamics within these groups compare.To investigate this, we monitored both direct and indirect subpopulations as two macaque monkeys learned to control a BMI. We found that while the combined population increased coordinated neural dynamics, this increase in coordination was primarily driven by changes in the direct subpopulation. These findings suggest that motor cortex refines cortical dynamics by increasing neural variance throughout the entire population during learning, with a more pronounced coordination of firing activity in subpopulations that are causally linked to behavior.

Modulation of neural co-firing to enhance network transmission and improve motor function after stroke

Stroke is a leading cause of disability. While neurotechnology has shown promise for improving upper limb recovery after stroke, efficacy in clinical trials has been variable. Our central thesis is that to improve clinical translation, we need to develop a common neurophysiological framework for understanding how neurotechnology alters network activity. Our perspective discusses principles for how motor networks, both healthy and those recovering from stroke, subserve reach-to-grasp movements. We focus on neural processing at the resolution of single movements, the timescale at which neurotechnologies are applied, and discuss how this activity might drive long-term plasticity. We propose that future studies should focus on cross-area communication and bridging our understanding of timescales ranging from single trials within a session to across multiple sessions. We hope that this perspective establishes a combined path forward for preclinical and clinical research with the goal of more robust clinical translation of neurotechnology.

Vagus nerve stimulation drives selective circuit modulation through cholinergic reinforcement

Vagus nerve stimulation (VNS) is a neuromodulation therapy for a broad and expanding set of neurologic conditions. However, the mechanism through which VNS influences central nervous system circuitry is not well described, limiting therapeutic optimization. VNS leads to widespread brain activation, but the effects on behavior are remarkably specific, indicating plasticity unique to behaviorally engaged neural circuits. To understand how VNS can lead to specific circuit modulation, we leveraged genetic tools including optogenetics and in vivo calcium imaging in mice learning a skilled reach task. We find that VNS enhances skilled motor learning in healthy animals via a cholinergic reinforcement mechanism, producing a rapid consolidation of an expert reach trajectory. In primary motor cortex (M1), VNS drives precise temporal modulation of neurons that respond to behavioral outcome. This suggests that VNS may accelerate motor refinement in M1 via cholinergic signaling, opening new avenues for optimizing VNS to target specific disease-relevant circuitry.

Modeling Neural Connectivity in a Point-Process Analogue of Kalman Filter

A neural encoding model describes how single neuron tunes to external stimuli as well as its connectivity with other neurons. The connectivity illustrates the neuronal interaction within populations in response to the shared latent brain states. Understanding these interactions is crucial to computationally predict the neural activity, which elucidates the neural encoding mechanism A computational analysis on the neural connectivity also facilitates developing more point process decoding model to interpret movement state from neural spike observations for brain machine interfaces (BMI). Most of the previous point process models only consider single neural tuning property and assumes conditional independence among multiple neurons. The connectivity among neurons is not considered in such a Bayesian approach to derive the state. In this work, we propose a point-process analogue of Kalman Filter to model the neural connectivity in a closed-form Bayesian filter. Neural connectivity corrects the posterior of the state given the multi-dimension observation, and a Gaussian distribution is used to approximate the updated posterior distribution. We validate the proposed method on simulation data and compared with traditional point process filtering with conditional independent assumption. The result shows that our method models the neural connectivity information and the single neuronal tuning property at the same time and achieves a better performance of the state decoding. Clinical Relevance - This paper proposes a closed-form derivation of a point process filter based on Gaussian approximations. It can model both single neuronal tuning property and the neural connectivity, which is potential to understanding the neural connectivity computationally.

From Parametric Representation to Dynamical System: Shifting Views of the Motor Cortex in Motor Control

In contrast to traditional representational perspectives in which the motor cortex is involved in motor control via neuronal preference for kinetics and kinematics, a dynamical system perspective emerging in the last decade views the motor cortex as a dynamical machine that generates motor commands by autonomous temporal evolution. In this review, we first look back at the history of the representational and dynamical perspectives and discuss their explanatory power and controversy from both empirical and computational points of view. Here, we aim to reconcile the above perspectives, and evaluate their theoretical impact, future direction, and potential applications in brain-machine interfaces.

Investigation of Neural Substrates of Erroneous Behavior in a Delayed-Response Task

Motor cortical neurons exhibit persistent selective activities (selectivity) during motor planning. Experimental perturbation of selectivity results in the failure of short-term memory retention and consequent behavioral biases, demonstrating selectivity as a neural characteristic of encoding previous sensory input or future action. However, even without experimental manipulation, animals occasionally fail to maintain short-term memory leading to erroneous choice. Here, we investigated neural substrates that lead to the incorrect formation of selectivity during short-term memory. We analyzed neuronal activities in anterior lateral motor cortex (ALM) of mice, a region known to be engaged in motor planning while mice performed the tactile delayed-response task. We found that highly selective neurons lost their selectivity while originally nonselective neurons showed selectivity during the error trials where mice licked toward incorrect direction. We assumed that those alternations would reflect changes in intrinsic properties of population activity. Thus, we estimated an intrinsic manifold shared by neuronal population (shared space), using factor analysis (FA) and measured the association of individual neurons with the shared space by communality, the variance of neuronal activity accounted for by the shared space. We found a positive correlation between selectivity and communality over ALM neurons, which disappeared in erroneous behavior. Notably, neurons showing selectivity alternations between correct and incorrect licking also underwent proportional changes in communality. Our results demonstrated that the extent to which an ALM neuron is associated with the intrinsic manifolds of population activity may elucidate its selectivity and that disruption of this association may alter selectivity, likely leading to erroneous behavior.

Reward and plasticity: Implications for neurorehabilitation

Neuroplasticity follows nervous system injury in the presence or absence of rehabilitative treatments. Rehabilitative interventions can be used to modulate adaptive neuroplasticity, reducing motor impairment and improving activities of daily living in patients with brain lesions. Learning principles guide some rehabilitative interventions. While basic science research has shown that reward combined with training enhances learning, this principle has been only recently explored in the context of neurorehabilitation. Commonly used reinforcers may be more or less rewarding depending on the individual or the context in which the task is performed. Studies in healthy humans showed that both reward and punishment can enhance within-session motor performance; but reward, and not punishment, improves consolidation and retention of motor skills. On the other hand, neurorehabilitative training after brain lesions involves complex tasks (e.g., walking and activities of daily living). The contribution of reward to neurorehabilitation is incompletely understood. Here, we discuss recent research on the role of reward in neurorehabilitation and the needed directions of future research.

Compartmentalized dynamics within a common multi-area mesoscale manifold represent a repertoire of human hand movements

The human hand is unique in the animal kingdom for unparalleled dexterity, ranging from complex prehension to fine finger individuation. How does the brain represent such a diverse repertoire of movements? We evaluated mesoscale neural dynamics across the human "grasp network," using electrocorticography and dimensionality reduction methods, for a repertoire of hand movements. Strikingly, we found that the grasp network represented both finger and grasping movements alike. Specifically, the manifold characterizing the multi-areal neural covariance structure was preserved during all movements across this distributed network. In contrast, latent neural dynamics within this manifold were surprisingly specific to movement type. Aligning latent activity to kinematics further uncovered distinct submanifolds despite similarities in synergistic coupling of joints between movements. We thus find that despite preserved neural covariance at the distributed network level, mesoscale dynamics are compartmentalized into movement-specific submanifolds; this mesoscale organization may allow flexible switching between a repertoire of hand movements.

Timescales of Local and Cross-Area Interactions during Neuroprosthetic Learning

How does the brain integrate signals with different timescales to drive purposeful actions? Brain-machine interfaces (BMIs) offer a powerful tool to causally test how distributed neural networks achieve specific neural patterns. During neuroprosthetic learning, actuator movements are causally linked to primary motor cortex (M1) neurons, i.e., "direct" neurons that project to the decoder and whose firing is required to successfully perform the task. However, it is unknown how such direct M1 neurons interact with both "indirect" local (in M1 but not part of the decoder) and across area neural populations (e.g., in premotor cortex/M2), all of which are embedded in complex biological recurrent networks. Here, we trained male rats to perform a M1-BMI task and simultaneously recorded the activity of indirect neurons in both M2 and M1. We found that both M2 and M1 indirect neuron populations could be used to predict the activity of the direct neurons (i.e., "BMI-potent activity"). Interestingly, compared with M1 indirect activity, M2 neural activity was correlated with BMI-potent activity across a longer set of time lags, and the timescale of population activity patterns evolved more slowly. M2 units also predicted the activity of both M1 direct and indirect neural populations, suggesting that M2 population dynamics provide a continuous modulatory influence on M1 activity as a whole, rather than a moment-by-moment influence solely on neurons most relevant to a task. Together, our results indicate that longer timescale M2 activity provides modulatory influence over extended time lags on shorter-timescale control signals in M1.SIGNIFICANCE STATEMENT A central question in the study of motor control is whether primary motor cortex (M1) and premotor cortex (M2) interact through task-specific subpopulations of neurons, or whether tasks engage broader correlated networks. Brain-machine interfaces (BMIs) are powerful tools to study cross-area interactions. Here, we performed simultaneous recordings of M1 and M2 in a BMI task using a subpopulation of M1 neurons (direct neurons). We found that activity outside of direct neurons in M1 and M2 was predictive of M1-BMI task activity, and that M2 activity evolved at slower timescales than M1. These findings suggest that M2 provides a continuous modulatory influence on M1 as a whole, supporting a model of interactions through broad correlated networks rather than task-specific neural subpopulations.

Hybrid dedicated and distributed coding in PMd/M1 provides separation and interaction of bilateral arm signals

Pronounced activity is observed in both hemispheres of the motor cortex during preparation and execution of unimanual movements. The organizational principles of bi-hemispheric signals and the functions they serve throughout motor planning remain unclear. Using an instructed-delay reaching task in monkeys, we identified two components in population responses spanning PMd and M1. A “dedicated” component, which segregated activity at the level of individual units, emerged in PMd during preparation. It was most prominent following movement when M1 became strongly engaged, and principally involved the contralateral hemisphere. In contrast to recent reports, these dedicated signals solely accounted for divergence of arm-specific neural subspaces. The other “distributed” component mixed signals for each arm within units, and the subspace containing it did not discriminate between arms at any stage. The statistics of the population response suggest two functional aspects of the cortical network: one that spans both hemispheres for supporting preparatory and ongoing processes, and another that is predominantly housed in the contralateral hemisphere and specifies unilateral output.

The motor cortex of the brain primarily controls the opposite side of the body, yet neural activity in this area is often observed during movements of either arm. To understand the functional significance of these signals we must first characterize how they are organized across the neural network. Are there patterns of activity that are unique to a single arm? Are there other patterns that reflect shared functions? Importantly, these features may change across time as motor plans are developed and executed. In this study, we analyzed the responses of individual neurons in the motor cortex and modeled their patterns of co-activity across the population to characterize the changes that distinguish left and right arm use. Across preparation and execution phases of the task, we found that signals became gradually more segregated. Despite many neurons modulating in association with either arm, those that were more dedicated to a single (typically contralateral) limb accounted for a disproportionately large amount of the variance. However, there were also weaker patterns of activity that did not distinguish between the two arms at any stage. These results reveal a heterogeneity in the motor cortex that highlights both independent and interactive components of reaching signals.

Changes in Modulation Characteristics of Neurons in Different Modes of Motion Control Using Brain-Machine Interface

In the research of motion control using brain-machine interface (BMI), analysis is usually conducted on one ensemble of neurons whose activity serves as direct input to the BMI decoder (control units). The number of control units is diverse in different control modes. That is to say, the size of dimensions of neural signals used in motion control is diverse. However, how will the behavioral performance change with this kind of diversity? What effects does this diversity have on modulation characteristics of control units? To answer these questions, we designed three modes of motion tasks using neural signals with different dimension sizes to control. Our results imply that as the dimension reduces, some deviations appear in behavioral performance. At the same time, the control units tend to have a directional division of control, then enhance their stability and increase modulations after division.

Bridging neuronal correlations and dimensionality reduction

Two commonly used approaches to study interactions among neurons are spike count correlation, which describes pairs of neurons, and dimensionality reduction, applied to a population of neurons. Although both approaches have been used to study trial-to-trial neuronal variability correlated among neurons, they are often used in isolation and have not been directly related. We first established concrete mathematical and empirical relationships between pairwise correlation and metrics of population-wide covariability based on dimensionality reduction. Applying these insights to macaque V4 population recordings, we found that the previously reported decrease in mean pairwise correlation associated with attention stemmed from three distinct changes in population-wide covariability. Overall, our work builds the intuition and formalism to bridge between pairwise correlation and population-wide covariability and presents a cautionary tale about the inferences one can make about population activity by using a single statistic, whether it be mean pairwise correlation or dimensionality.

Computation Through Neural Population Dynamics

Significant experimental, computational, and theoretical work has identified rich structure within the coordinated activity of interconnected neural populations. An emerging challenge now is to uncover the nature of the associated computations, how they are implemented, and what role they play in driving behavior. We term this computation through neural population dynamics. If successful, this framework will reveal general motifs of neural population activity and quantitatively describe how neural population dynamics implement computations necessary for driving goal-directed behavior. Here, we start with a mathematical primer on dynamical systems theory and analytical tools necessary to apply this perspective to experimental data. Next, we highlight some recent discoveries resulting from successful application of dynamical systems. We focus on studies spanning motor control, timing, decision-making, and working memory. Finally, we briefly discuss promising recent lines of investigation and future directions for the computation through neural population dynamics framework.

Strengths and challenges of longitudinal non-human primate neuroimaging

Longitudinal non-human primate neuroimaging has the potential to greatly enhance our understanding of primate brain structure and function. Here we describe its specific strengths, compared to both cross-sectional non-human primate neuroimaging and longitudinal human neuroimaging, but also its associated challenges. We elaborate on factors guiding the use of different analytical tools, subject-specific versus age-specific templates for analyses, and issues related to statistical power.

Learning is shaped by abrupt changes in neural engagement

Internal states such as arousal, attention and motivation modulate brain-wide neural activity, but how these processes interact with learning is not well understood. During learning, the brain modifies its neural activity to improve behavior. How do internal states affect this process? Using a brain-computer interface learning paradigm in monkeys, we identified large, abrupt fluctuations in neural population activity in motor cortex indicative of arousal-like internal state changes, which we term 'neural engagement.' In a brain-computer interface, the causal relationship between neural activity and behavior is known, allowing us to understand how neural engagement impacted behavioral performance for different task goals. We observed stereotyped changes in neural engagement that occurred regardless of how they impacted performance. This allowed us to predict how quickly different task goals were learned. These results suggest that changes in internal states, even those seemingly unrelated to goal-seeking behavior, can systematically influence how behavior improves with learning.

The sensory representation of causally controlled objects

Intentional control over external objects is informed by our sensory experience of them. To study how causal relationships are learned and effected, we devised a brain machine interface (BMI) task using wide-field calcium signals. Mice learned to entrain activity patterns in arbitrary pairs of cortical regions to guide a visual cursor to a target location for reward. Brain areas that were normally correlated could be rapidly reconfigured to exert control over the cursor in a sensory-feedback-dependent manner. Higher visual cortex was more engaged when expert but not naive animals controlled the cursor. Individual neurons in higher visual cortex responded more strongly to the cursor when mice controlled it than when they passively viewed it, with the greatest response boosting as the cursor approached the target location. Thus, representations of causally controlled objects are sensitive to intention and proximity to the subject's goal, potentially strengthening sensory feedback to allow more fluent control.

Low-frequency stimulation enhances ensemble co-firing and dexterity after stroke

Electrical stimulation is a promising tool for modulating brain networks. However, it is unclear how stimulation interacts with neural patterns underlying behavior. Specifically, how might external stimulation that is not sensitive to the state of ongoing neural dynamics reliably augment neural processing and improve function? Here, we tested how low-frequency epidural alternating current stimulation (ACS) in non-human primates recovering from stroke interacted with task-related activity in perilesional cortex and affected grasping. We found that ACS increased co-firing within task-related ensembles and improved dexterity. Using a neural network model, we found that simulated ACS drove ensemble co-firing and enhanced propagation of neural activity through parts of the network with impaired connectivity, suggesting a mechanism to link increased co-firing to enhanced dexterity. Together, our results demonstrate that ACS restores neural processing in impaired networks and improves dexterity following stroke. More broadly, these results demonstrate approaches to optimize stimulation to target neural dynamics.

Encouraging an excitable brain state: mechanisms of brain repair in stroke

Stroke induces a plastic state in the brain. This period of enhanced plasticity leads to the sprouting of new axons, the formation of new synapses and the remapping of sensory-motor functions, and is associated with motor recovery. This is a remarkable process in the adult brain, which is normally constrained in its levels of neuronal plasticity and connectional change. Recent evidence indicates that these changes are driven by molecular systems that underlie learning and memory, such as changes in cellular excitability during memory formation. This Review examines circuit changes after stroke, the shared mechanisms between memory formation and brain repair, the changes in neuronal excitability that underlie stroke recovery, and the molecular and pharmacological interventions that follow from these findings to promote motor recovery in animal models. From these findings, a framework emerges for understanding recovery after stroke, central to which is the concept of neuronal allocation to damaged circuits. The translation of the concepts discussed here to recovery in humans is underway in clinical trials for stroke recovery drugs.

Adenosine A2A receptor blockade improves neuroprosthetic learning by volitional control of population calcium signal in M1 cortical neurons

Volitional control is at the core of brain-machine interfaces (BMI) adaptation and neuroprosthetic-driven learning to restore motor function for disabled patients, but neuroplasticity changes and neuromodulation underlying volitional control of neuroprosthetic learning are largely unexplored. To better study volitional control at annotated neural population, we have developed an operant neuroprosthetic task with closed-loop feedback system by volitional conditioning of population calcium signal in the M1 cortex using fiber photometry recording. Importantly, volitional conditioning of the population calcium signal in M1 neurons did not improve within-session adaptation, but specifically enhanced across-session neuroprosthetic skill learning with reduced time-to-target and the time to complete 50 successful trials. With brain-behavior causality of the neuroprosthetic paradigm, we revealed that proficiency of neuroprosthetic learning by volitional conditioning of calcium signal was associated with the stable representational (plasticity) mapping in M1 neurons with the reduced calcium peak. Furthermore, pharmacological blockade of adenosine A_2A receptors facilitated volitional conditioning of neuroprosthetic learning and converted an ineffective volitional conditioning protocol to be the effective for neuroprosthetic learning. These findings may help to harness neuroplasticity for better volitional control of neuroprosthetic training and suggest a novel pharmacological strategy to improve neuroprosthetic learning in BMI adaptation by targeting striatal A_2A receptors.

Neuronal Activity Distributed in Multiple Cortical Areas during Voluntary Control of the Native Arm or a Brain-Computer Interface

Voluntary control of visually-guided upper extremity movements involves neuronal activity in multiple areas of the cerebral cortex. Studies of brain-computer interfaces (BCIs) that use spike recordings for input, however, have focused largely on activity in the region from which those neurons that directly control the BCI, which we call BCI units, are recorded. We hypothesized that just as voluntary control of the arm and hand involves activity in multiple cortical areas, so does voluntary control of a BCI. In two subjects (Macaca mulatta) performing a center-out task both with a hand-held joystick and with a BCI directly controlled by four primary motor cortex (M1) BCI units, we recorded the activity of other, non-BCI units in M1, dorsal premotor cortex (PMd) and ventral premotor cortex (PMv), primary somatosensory cortex (S1), dorsal posterior parietal cortex (dPPC), and the anterior intraparietal area (AIP). In most of these areas, non-BCI units were active in similar percentages and at similar modulation depths during both joystick and BCI trials. Both BCI and non-BCI units showed changes in preferred direction (PD). Additionally, the prevalence of effective connectivity between BCI and non-BCI units was similar during both tasks. The subject with better BCI performance showed increased percentages of modulated non-BCI units with increased modulation depth and increased effective connectivity during BCI as compared with joystick trials; such increases were not found in the subject with poorer BCI performance. During voluntary, closed-loop control, non-BCI units in a given cortical area may function similarly whether the effector is the native upper extremity or a BCI-controlled device.

Orthogonalizing the Activity of Two Neural Units for 2D Cursor Movement Control

In the design of brain-machine interface (BMI), as the number of electrodes used to collect neural spike signals declines slowly, it is important to be able to decode with fewer units. We tried to train a monkey to control a cursor to perform a two-dimensional (2D) center-out task smoothly with spiking activities only from two units (direct units). At the same time, we studied how the direct units did change their tuning to the preferred direction during BMI training and tried to explore the underlying mechanism of how the monkey learned to control the cursor with their neural signals. In this study, we observed that both direct units slowly changed their preferred directions during BMI learning. Although the initial angles between the preferred directions of 3 pairs units are different, the angle between their preferred directions approached 90 degrees at the end of the training. Our results imply that BMI learning made the two units independent of each other. To our knowledge, it is the first time to demonstrate that only two units could be used to control a 2D cursor movements. Meanwhile, orthogonalizing the activities of two units driven by BMI learning in this study implies that the plasticity of the motor cortex is capable of providing an efficient strategy for motor control.

Single-trial cross-area neural population dynamics during long-term skill learning

Mammalian cortex has both local and cross-area connections, suggesting vital roles for both local and cross-area neural population dynamics in cortically-dependent tasks, like movement learning. Prior studies of movement learning have focused on how single-area population dynamics change during short-term adaptation. It is unclear how cross-area dynamics contribute to movement learning, particularly long-term learning and skill acquisition. Using simultaneous recordings of rodent motor (M1) and premotor (M2) cortex and computational methods, we show how cross-area activity patterns evolve during reach-to-grasp learning in rats. The emergence of reach-related modulation in cross-area activity correlates with skill acquisition, and single-trial modulation in cross-area activity predicts reaction time and reach duration. Local M2 neural activity precedes local M1 activity, supporting top-down hierarchy between the regions. M2 inactivation preferentially affects cross-area dynamics and behavior, with minimal disruption of local M1 dynamics. Together, these results indicate that cross-area population dynamics are necessary for learned motor skills.

Reconfiguring Motor Circuits for a Joint Manual and BCI Task

Designing brain-computer interfaces (BCIs) that can be used in conjunction with ongoing motor behavior requires an understanding of how neural activity co-opted for brain control interacts with existing neural circuits. For example, BCIs may be used to regain lost motor function after stroke. This requires that neural activity controlling unaffected limbs is dissociated from activity controlling the BCI. In this study we investigated how primary motor cortex accomplishes simultaneous BCI control and motor control in a task that explicitly required both activities to be driven from the same brain region (i.e. a dual-control task). Single-unit activity was recorded from intracortical, multi-electrode arrays while a non-human primate performed this dual-control task. Compared to activity observed during naturalistic motor control, we found that both units used to drive the BCI directly (control units) and units that did not directly control the BCI (non-control units) significantly changed their tuning to wrist torque. Using a measure of effective connectivity, we observed that control units decrease their connectivity. Through an analysis of variance we found that the intrinsic variability of the control units has a significant effect on task proficiency. When this variance is accounted for, motor cortical activity is flexible enough to perform novel BCI tasks that require active decoupling of natural associations to wrist motion. This study provides insight into the neural activity that enables a dual-control brain-computer interface.

Neural reinforcement: re-entering and refining neural dynamics leading to desirable outcomes

How do organisms learn to do again, on-demand, a behavior that led to a desirable outcome? Dopamine-dependent cortico-striatal plasticity provides a framework for learning behavior's value, but it is less clear how it enables the brain to re-enter desired behaviors and refine them over time. Reinforcing behavior is achieved by re-entering and refining the neural patterns that produce it. We review studies using brain-machine interfaces which reveal that reinforcing cortical population activity requires cortico-basal ganglia circuits. Then, we propose a formal framework for how reinforcement in cortico-basal ganglia circuits acts on the neural dynamics of cortical populations. We propose two parallel mechanisms: i) fast reinforcement which selects the inputs that permit the re-entrance of the particular cortical population dynamics which naturally produced the desired behavior, and ii) slower reinforcement which leads to refinement of cortical population dynamics and more reliable production of neural trajectories driving skillful behavior on-demand.

New neural activity patterns emerge with long-term learning

Learning has been associated with changes in the brain at every level of organization. However, it remains difficult to establish a causal link between specific changes in the brain and new behavioral abilities. We establish that new neural activity patterns emerge with learning. We demonstrate that these new neural activity patterns cause the new behavior. Thus, the formation of new patterns of neural population activity can underlie the learning of new skills.

Data-driven models in human neuroscience and neuroengineering

Discoveries in modern human neuroscience are increasingly driven by quantitative understanding of complex data. Data-intensive approaches to modeling have promise to dramatically advance our understanding of the brain and critically enable neuroengineering capabilities. In this review, we provide an accessible primer to modern modeling approaches and highlight recent data-driven discoveries in the domains of neuroimaging, single-neuron and neuronal population responses, and device neuroengineering. Further, we suggest that meaningful progress requires the community to tackle open challenges in the realms of model interpretability and generalizability, training pipelines of data-fluent human neuroscientists, and integrated consideration of data ethics.

Bayesian Computation through Cortical Latent Dynamics

Statistical regularities in the environment create prior beliefs that we rely on to optimize our behavior when sensory information is uncertain. Bayesian theory formalizes how prior beliefs can be leveraged and has had a major impact on models of perception, sensorimotor function, and cognition. However, it is not known how recurrent interactions among neurons mediate Bayesian integration. By using a time-interval reproduction task in monkeys, we found that prior statistics warp neural representations in the frontal cortex, allowing the mapping of sensory inputs to motor outputs to incorporate prior statistics in accordance with Bayesian inference. Analysis of recurrent neural network models performing the task revealed that this warping was enabled by a low-dimensional curved manifold and allowed us to further probe the potential causal underpinnings of this computational strategy. These results uncover a simple and general principle whereby prior beliefs exert their influence on behavior by sculpting cortical latent dynamics.

Emerging ideas and tools to study the emergent properties of the cortical neural circuits for voluntary motor control in non-human primates

For years, neurophysiological studies of the cerebral cortical mechanisms of voluntary motor control were limited to single-electrode recordings of the activity of one or a few neurons at a time. This approach was supported by the widely accepted belief that single neurons were the fundamental computational units of the brain (the “neuron doctrine”). Experiments were guided by motor-control models that proposed that the motor system attempted to plan and control specific parameters of a desired action, such as the direction, speed or causal forces of a reaching movement in specific coordinate frameworks, and that assumed that the controlled parameters would be expressed in the task-related activity of single neurons. The advent of chronically implanted multi-electrode arrays about 20 years ago permitted the simultaneous recording of the activity of many neurons. This greatly enhanced the ability to study neural control mechanisms at the population level. It has also shifted the focus of the analysis of neural activity from quantifying single-neuron correlates with different movement parameters to probing the structure of multi-neuron activity patterns to identify the emergent computational properties of cortical neural circuits. In particular, recent advances in “dimension reduction” algorithms have attempted to identify specific covariance patterns in multi-neuron activity which are presumed to reflect the underlying computational processes by which neural circuits convert the intention to perform a particular movement into the required causal descending motor commands. These analyses have led to many new perspectives and insights on how cortical motor circuits covertly plan and prepare to initiate a movement without causing muscle contractions, transition from preparation to overt execution of the desired movement, generate muscle-centered motor output commands, and learn new motor skills. Progress is also being made to import optical-imaging and optogenetic toolboxes from rodents to non-human primates to overcome some technical limitations of multi-electrode recording technology.

Brain-Machine Interfaces: Powerful Tools for Clinical Treatment and Neuroscientific Investigations

Brain-machine interfaces (BMIs) have exploded in popularity in the past decade. BMIs, also called brain-computer interfaces, provide a direct link between the brain and a computer, usually to control an external device. BMIs have a wide array of potential clinical applications, ranging from restoring communication to people unable to speak due to amyotrophic lateral sclerosis or a stroke, to restoring movement to people with paralysis from spinal cord injury or motor neuron disease, to restoring memory to people with cognitive impairment. Because BMIs are controlled directly by the activity of prespecified neurons or cortical areas, they also provide a powerful paradigm with which to investigate fundamental questions about brain physiology, including neuronal behavior, learning, and the role of oscillations. This article reviews the clinical and neuroscientific applications of BMIs, with a primary focus on motor BMIs.

Distinct types of neural reorganization during long-term learning

What are the neural mechanisms of skill acquisition? Many studies find that long-term practice is associated with a functional reorganization of cortical neural activity. However, the link between these changes in neural activity and the behavioral improvements that occur is not well understood, especially for long-term learning that takes place over several weeks. To probe this link in detail, we leveraged a brain-computer interface (BCI) paradigm in which rhesus monkeys learned to master nonintuitive mappings between neural spiking in primary motor cortex and computer cursor movement. Critically, these BCI mappings were designed to disambiguate several different possible types of neural reorganization. We found that during the initial phase of learning, lasting minutes to hours, rapid changes in neural activity common to all neurons led to a fast suppression of motor error. In parallel, local changes to individual neurons gradually accrued over several weeks of training. This slower timescale cortical reorganization persisted long after the movement errors had decreased to asymptote and was associated with more efficient control of movement. We conclude that long-term practice evokes two distinct neural reorganization processes with vastly different timescales, leading to different aspects of improvement in motor behavior. NEW & NOTEWORTHY We leveraged a brain-computer interface learning paradigm to track the neural reorganization occurring throughout the full time course of motor skill learning lasting several weeks. We report on two distinct types of neural reorganization that mirror distinct phases of behavioral improvement: a fast phase, in which global reorganization of neural recruitment leads to a quick suppression of motor error, and a slow phase, in which local changes in individual tuning lead to improvements in movement efficiency.

Latent Factors and Dynamics in Motor Cortex and Their Application to Brain-Machine Interfaces

In the 1960s, Evarts first recorded the activity of single neurons in motor cortex of behaving monkeys (Evarts, 1968). In the 50 years since, great effort has been devoted to understanding how single neuron activity relates to movement. Yet these single neurons exist within a vast network, the nature of which has been largely inaccessible. With advances in recording technologies, algorithms, and computational power, the ability to study these networks is increasing exponentially. Recent experimental results suggest that the dynamical properties of these networks are critical to movement planning and execution. Here we discuss this dynamical systems perspective and how it is reshaping our understanding of the motor cortices. Following an overview of key studies in motor cortex, we discuss techniques to uncover the "latent factors" underlying observed neural population activity. Finally, we discuss efforts to use these factors to improve the performance of brain-machine interfaces, promising to make these findings broadly relevant to neuroengineering as well as systems neuroscience.

Emergent coordination with a brain-machine interface: implications for the neural basis of motor learning

How patterns of covariance in motor output and neural activity emerge over the course of learning is a topic of ongoing investigation. Vaidya et al. (Vaidya M, Balasubramanian K, Southerland J, Badreldin I, Eleryan A, Shattuck K, Gururangan S, Slutzky M, Osborne L, Fagg A, Oweiss K, Hatsopoulos NG. J Neurophysiol 119: 1291-1304, 2018) investigate the emergence of patterns of covariance in the motor output and neural activity in chronically amputated macaques learning reach-to-grasp movements with a brain-machine interface. The authors' findings have implications for uncovering general principles of how neural coordination unfolds while learning a different motor behavior.