Tuning Curves for Arm Posture Control in Motor Cortex Are Consistent with Random Connectivity.

Neuronal responses characterized by regular tuning curves are typically assumed to arise from structured synaptic connectivity. However, many responses exhibit both regular and irregular components. To address the relationship between tuning curve properties and underlying circuitry, we analyzed neuronal activity recorded from primary motor cortex (M1) of monkeys performing a 3D arm posture control task and compared the results with a neural network model. Posture control is well suited for examining M1 neuronal tuning because it avoids the dynamic complexity of time-varying movements. As a function of hand position, the neuronal responses have a linear component, as has previously been described, as well as heterogeneous and highly irregular nonlinearities. These nonlinear components involve high spatial frequencies and therefore do not support explicit encoding of movement parameters. Yet both the linear and nonlinear components contribute to the decoding of EMG of major muscles used in the task. Remarkably, despite the presence of a strong linear component, a feedforward neural network model with entirely random connectivity can replicate the data, including both the mean and distributions of the linear and nonlinear components as well as several other features of the neuronal responses. This result shows that smoothness provided by the regularity in the inputs to M1 can impose apparent structure on neural responses, in this case a strong linear (also known as cosine) tuning component, even in the absence of ordered synaptic connectivity.

A Bayesian approach for characterizing direction tuning curves in the supplementary motor area of behaving monkeys.

Neural responses are commonly studied in terms of "tuning curves," characterizing changes in neuronal response as a function of a continuous stimulus parameter. In the motor system, neural responses to movement direction often follow a bell-shaped tuning curve for which the exact shape determines the properties of neuronal movement coding. Estimating the shape of that tuning curve robustly is hard, especially when directions are sampled unevenly and at a coarse resolution. Here, we describe a Bayesian estimation procedure that improves the accuracy of curve-shape estimation even when the curve is sampled unevenly and at a very coarse resolution. Using this approach, we characterize the movement direction tuning curves in the supplementary motor area (SMA) of behaving monkeys. We compare the SMA tuning curves to tuning curves of neurons from the primary motor cortex (M1) of the same monkeys, showing that the tuning curves of the SMA neurons tend to be narrower and shallower. We also show that these characteristics do not depend on the specific location in each region.

The representation of visual and motor aspects of reaching movements in the human motor cortex.

The human primary motor cortex (M1) is robustly activated during visually guided hand movements. M1 multivoxel patterns of functional MRI activation are more correlated during repeated hand movements to the same targets than to greatly differing ones, and therefore potentially contain information about movement direction. It is unclear, however, whether direction specificity is due to the motor command, as implicitly assumed, or to the visual aspects of the task, such as the target location and the direction of the cursor's trajectory. To disambiguate the visual and motor components, different visual-to-motor transformations were applied during an fMRI scan, in which participants made visually guided hand movements in various directions. The first run was the "baseline" (i.e., visual and motor mappings were matched); in the second run ("rotation"), the cursor movement was rotated by 45° with respect to the joystick movement. As expected, positive correlations were seen between the M1 multivoxel patterns evoked by the baseline run and by the rotation run, when the two movements were matched in their movement direction but the visual aspects differed. Importantly, similar correlations were observed when the visual elements were matched but the direction of hand movement differed. This indicates that M1 is sensitive to both motor and visual components of the task. However, repeated observation of the cursor movement without concurrent joystick control did not elicit significant activation in M1 or any correlated patterns of activation. Thus, visual aspects of movement are encoded in M1 only when they are coupled with motor consequences.

The neuronal basis of long-term sensorimotor learning.

The brain has a remarkable ability to learn and adjust behavior. For instance, the brain can adjust muscle activation to cope with changes in the environment. However, the neuronal mechanisms behind this adaptation are not clear. To address this fundamental question, this study examines the neuronal basis of long-term sensorimotor learning by recording neuronal activity in the primary motor cortex of monkeys during a long-term adaptation to a force-field perturbation. For 5 consecutive days, the same perturbation was applied to the monkey's hand when reaching to a single target, whereas movements to all other targets were not perturbed. The gradual improvement in performance over these 5 days was correlated to the evolvement in the population neuronal signal, with two timescales of changes in single-cell activity. Specifically, one subgroup of cells showed a relatively fast increase in activity, whereas the other showed a gradual, slower decrease. These adapted patterns of neuronal activity did not involve changes in directional tuning of single cells, suggesting that adaptation was the result of adjustments of the required motor plan by a population of neurons rather than changes in single-cell properties. Furthermore, generalization was mostly expressed in the direction of the required compensatory force during adaptation. Altogether, the neuronal activity and its generalization accord with the adapted motor plan.

The minimum information principle and its application to neural code analysis.

The study of complex information processing systems requires appropriate theoretical tools to help unravel their underlying design principles. Information theory is one such tool, and has been utilized extensively in the study of the neural code. Although much progress has been made in information theoretic methodology, there is still no satisfying answer to the question: "What is the information that a given property of the neural population activity (e.g., the responses of single cells within the population) carries about a set of stimuli?" Here, we answer such questions via the minimum mutual information (MinMI) principle. We quantify the information in any statistical property of the neural response by considering all hypothetical neuronal populations that have the given property and finding the one that contains the minimum information about the stimuli. All systems with higher information values necessarily contain additional information processing mechanisms and, thus, the minimum captures the information related to the given property alone. MinMI may be used to measure information in properties of the neural response, such as that conveyed by responses of small subsets of cells (e.g., singles or pairs) in a large population and cooperative effects between subunits in networks. We show how the framework can be used to study neural coding in large populations and to reveal properties that are not discovered by other information theoretic methods.

Functional organization of human motor cortex: directional selectivity for movement.

In monkeys, neurons in the hand representation of the primary motor cortex (M1) are often tuned to the direction of hand movement, and there is evidence that these neurons are clustered according to their "preferred" direction of movement. However, this organizational principle has yet to be demonstrated in M1 of humans. We conducted a functional magnetic resonance imaging (fMRI) study in which participants used a joystick to move a cursor from a central origin to one of five equidistant targets. The fMRI signal of individual voxels was sensitive to the directional aspects of the reaching task and manifested direction-specific adaptation. Furthermore, the correlation between multivoxel patterns of responses for different movement directions depended on the angular distance between them. We conclude that M1 neurons are likely to be organized in clusters according to their preferred direction, since only such a coarse-grained representation can lead to directional selectivity of voxels, encompassing millions of neurons. A simple model that estimates cluster size suggests that the diameter of these clusters is on the order of a few hundred micrometers.

Neuronal correlates of memory formation in motor cortex after adaptation to force field.

Activity of single neurons in the motor cortex has been shown to change during acquisition of motor skills. We previously reported that the combined activity of cell ensembles in the motor cortex of monkeys (Macaca fascicularis) evolves during adaptation to a novel force field perturbation to encode the direction of compensatory force when reaching to visual targets. We also showed that the population directional signal was altered by the available sensory feedback. Here, we examined whether traces of such activity would linger on to later constitute motor memories of the newly acquired skill and whether memory traces would differ depending on feedback. We found that motor-cortical cell ensembles retained features of their adaptive activity pattern in the absence of perturbation when reaching to both learned and unlearned targets. Moreover, the preferred directions of these cells rotated in the direction of force field while the entire population of cells produced no net rotation of preferred direction when returning to null-field reaches. Whereas the activity pattern and preferred direction rotations were comparable with and without visual feedback, changes in tuning amplitudes differed across feedback conditions. Last, savings in behavioral performance and neuronal activity during later reexposure to force field were apparent. Overall, the findings reflect a novel representation of motor memory by cell ensembles and indicate a putative role of the motor cortex in early acquisition of motor memory.

Combined adaptiveness of specific motor cortical ensembles underlies learning.

Learning motor skills entails adaptation of neural computations that can generate or modify associations between sensations and actions. Indeed, humans can use different strategies when adapting to dynamic loads depending on available sensory feedback. Here, we examined how neural activity in motor cortex was modified when monkeys made arm reaches to a visual target and locally adapted to curl force field with or without visual trajectory feedback. We found that firing rates of a large subpopulation of cells were consistently modulated depending on the distance of their preferred direction from the learned movement direction. The newly acquired activity followed a cosine-like function, with maximal increase in directions that opposed the perturbing force and decrease in opposite directions. As a result, the combined neuronal activity generated an adapted population vector. The results suggest that this could be achieved without changing the tuning properties of the cells. This population directional signal was however altered in the absence of visual feedback; while the cosine pattern of modulation was maintained, the population distributions of modulated cells differed across feedback consistent with the different trajectory shapes. Finally, we predicted generalization patterns of force-field learning based on the cosine-like modulation. These conformed to reported features of generalization in humans, suggesting that the generalization function was related to the observed rate modulations in the motor cortex. Overall, the findings suggest that the new combined activation of neuronal ensembles could underlie the change in the internal model of movement dynamics in a way that depends on available sensory feedback and chosen strategy.

Trial-to-trial variability of single cells in motor cortices is dynamically modified during visuomotor adaptation.

Neurons in all brain areas exhibit variability in their spiking activity. Although part of this variability can be considered as noise that is detrimental to information processing, recent findings indicate that variability can also be beneficial. In particular, it was suggested that variability in the motor system allows for exploration of possible motor states and therefore can facilitate learning and adaptation to new environments. Here, we provide evidence to support this idea by analyzing the variability of neurons in the primary motor cortex (M1) and in the supplementary motor area (SMA-proper) of monkeys adapting to new rotational visuomotor tasks. We found that trial-to-trial variability increased during learning and exhibited four main characteristics: (1) modulation occurred preferentially during a delay period when the target of movement was already known, but before movement onset; (2) variability returned to its initial levels toward the end of learning; (3) the increase in variability was more apparent in cells with preferred movement directions close to those experienced during learning; and (4) the increase in variability emerged at early phases of learning in the SMA, whereas in M1 behavior reached plateau levels of performance. These results are highly consistent with previous findings that showed similar trends in variability across a population of neurons. Together, the results strengthen the idea that single-cell variability can be much more than mere noise and may be an integral part of the underlying mechanism of sensorimotor learning.

Midbrain dopamine neurons encode decisions for future action.

Current models of the basal ganglia and dopamine neurons emphasize their role in reinforcement learning. However, the role of dopamine neurons in decision making is still unclear. We recorded from dopamine neurons in monkeys engaged in two types of trial: reference trials in an instructed-choice task and decision trials in a two-armed bandit decision task. We show that the activity of dopamine neurons in the decision setting is modulated according to the value of the upcoming action. Moreover, analysis of the probability matching strategy in the decision trials revealed that the dopamine population activity and not the reward during reference trials determines choice behavior. Because dopamine neurons do not have spatial or motor properties, we conclude that immediate decisions are likely to be generated elsewhere and conveyed to the dopamine neurons, which play a role in shaping long-term decision policy through dynamic modulation of the efficacy of basal ganglia synapses.

Phase-Specific Microstimulation Differentially Modulates Beta Oscillations and Affects Behavior.

It is widely accepted that Beta-band oscillations play a role in sensorimotor behavior. To further explore this role, we developed a hybrid platform to combine neural operant conditioning and phase-specific intracortical microstimulation (ICMS). We trained monkeys, implanted with 96 electrode arrays in the motor cortex, to volitionally enhance local field potential (LFP) Beta-band (20-30 Hz) activity at selected sites using a brain-machine interface. We find that Beta oscillations of LFP and single-unit spiking activity increase dramatically with brain-machine interface training and that pre-movement Beta power is anti-correlated with task performance. We also find that phase-specific ICMS modulates the power and phase of oscillations, shifting local networks between oscillatory and non-oscillatory states. Furthermore, ICMS induces phase-dependent effects in animal reaction times and success rates. These findings contribute to unraveling the functional role of cortical oscillations and to the future development of clinical tools for ameliorating abnormal neuronal activities in brain disease.

Emergence of novel representations in primary motor cortex and premotor neurons during associative learning.

Neurons in the motor areas of cortex play a key role in associating sensory instructions with movements. However, their ability to acquire and maintain representations of novel stimulus features, especially when these features are behaviorally relevant, remains unknown. We investigated neuronal changes in these areas during and after associative learning, by training monkeys on a novel reaching task that required associating target colors with movement directions. Before and after learning, the monkeys performed a well known center-out task. We found that during learning, up to 48% of the neurons developed learning-related responses, differentiating between the associative task and the center-out task, although movement kinematics were the same. After learning, on returning to the center-out task in which color was irrelevant, many of these neurons maintained their response to the associative task; they displayed novel sensitivity to the color of the target that was relevant during learning. These neuronal responses prevailed in both the primary motor cortex and the ventral and dorsal premotor cortices, without degrading the information that the neurons firing carried about movement direction. Our results show that motor cortical neurons can rapidly develop and maintain sensitivities to novel arbitrary sensory features such as color, when such features are behaviorally relevant.

Midbrain dopaminergic neurons and striatal cholinergic interneurons encode the difference between reward and aversive events at different epochs of probabilistic classical conditioning trials.

Midbrain dopaminergic neurons (DANs) typically increase their discharge rate in response to appetitive predictive cues and outcomes, whereas striatal cholinergic tonically active interneurons (TANs) decrease their rate. This may indicate that the activity of TANs and DANs is negatively correlated and that TANs can broaden the basal ganglia reinforcement teaching signal, for instance by encoding worse than predicted events. We studied the activity of 106 DANs and 180 TANs of two monkeys recorded during the performance of a classical conditioning task with cues predicting the probability of food, neutral, and air puff outcomes. DANs responded to all cues with elevations of discharge rate, whereas TANs depressed their discharge rate. Nevertheless, although dopaminergic responses to appetitive cues were larger than their responses to neutral or aversive cues, the TAN responses were more similar. Both TANs and DANs responded faster to an air puff than to a food outcome; however, DANs responded with a discharge elevation, whereas the TAN responses included major negative and positive deflections. Finally, food versus air puff omission was better encoded by TANs. In terms of the activity of single neurons with distinct responses to the different behavioral events, both DANs and TANs were more strongly modulated by reward than by aversive related events and better reflected the probability of reward than aversive outcome. Thus, TANs and DANs encode the task episodes differentially. The DANs encode mainly the cue and outcome delivery, whereas the TANs mainly encode outcome delivery and omission at termination of the behavioral trial episode.

Low-pass filter properties of basal ganglia cortical muscle loops in the normal and MPTP primate model of parkinsonism.

Oscillatory bursting activity is commonly found in the basal ganglia (BG) and the thalamus of the parkinsonian brain. The frequency of these oscillations is often similar to or higher than that of the parkinsonian tremor, but their relationship to the tremor and other parkinsonian symptoms is still under debate. We studied the frequency dependency of information transmission in the cortex-BG and cortex-periphery loops by recording simultaneously from multiple electrodes located in the arm-related primary motor cortex (MI) and in the globus pallidus (GP) of two vervet monkeys before and after 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) treatment and induction of parkinsonian symptoms. We mimicked the parkinsonian bursting oscillations by stimulating with 35 ms bursts given at different frequencies through microelectrodes located in MI or GP while recording the evoked neuronal and motor responses. In the normal state, microstimulation of MI or GP does not modulate the discharge rate in the other structure. However, the functional-connectivity between MI and GP is greatly enhanced after MPTP treatment. In the frequency domain, GP neurons usually responded equally to 1-15 Hz stimulation bursts in both states. In contrast, MI neurons demonstrated low-pass filter properties, with a cutoff frequency above 5 Hz for the MI stimulations, and below 5 Hz for the GP stimulations. Finally, muscle activation evoked by MI microstimulation was markedly attenuated at frequencies higher than 5 Hz. The low-pass properties of the pathways connecting GP to MI to muscles suggest that parkinsonian tremor is not directly driven by the BG 5-10 Hz burst oscillations despite their similar frequencies.

Learning from learning: what can visuomotor adaptations tell us about the neuronal representation of movement?

The use of sensorimotor adaptation and learning paradigms in psychophyical and electrophysiological experiments can help to shed light on two fundamental questions. First, what are the computations that control sensorimotor behavior and, second, what are the neuronal mechanisms and representations underlying newly learned sensorimotor skills? We describe experiments that combined behavioral and electrophysioloigcal techniques and discuss implication of the results to three main questions: How do neuronal primitives of representation affect performance and learning? Do pre-motor and primary motor cortices form a hierarchy of computation, with different roles during learning and motor performance? How do these different cortical areas and the representations of movement change during the different stages of learning and memory formation?

Coincident but distinct messages of midbrain dopamine and striatal tonically active neurons.

Midbrain dopamine and striatal tonically active neurons (TANs, presumed acetylcholine interneurons) signal behavioral significance of environmental events. Since striatal dopamine and acetylcholine affect plasticity of cortico-striatal transmission and are both crucial to learning, they may serve as teachers in the basal ganglia circuits. We recorded from both neuronal populations in monkeys performing a probabilistic instrumental conditioning task. Both neuronal types respond robustly to reward-related events. Although different events yielded responses with different latencies, the responses of the two populations coincided, indicating integration at the target level. Yet, while the dopamine neurons' response reflects mismatch between expectation and outcome in the positive domain, the TANs are invariant to reward predictability. Finally, TAN pairs are synchronized, compared to a minority of dopamine neuron pairs. We conclude that the striatal cholinergic and dopaminergic systems carry distinct messages by different means, which can be integrated differently to shape the basal ganglia responses to reward-related events.

Preparatory activity in motor cortex reflects learning of local visuomotor skills.

In humans, learning to produce correct visually guided movements to adapt to new sensorimotor conditions requires the formation of an internal model that represents the new transformation between visual input and the required motor command. When the new environment requires adaptation to directional errors, learning generalizes poorly to untrained locations and directions, indicating that such learning is local. Here we replicated these behavioral findings in rhesus monkeys using a visuomotor rotation task and simultaneously recorded neuronal activity. Specific changes in activity were observed only in a subpopulation of cells in the motor cortex with directional properties corresponding to the locally learned rotation. These changes adhered to the dynamics of behavior during learning and persisted between learning and relearning of the same rotation. These findings suggest a neural mechanism for the locality of newly acquired sensorimotor tasks and provide electrophysiological evidence for their retention in working memory.

Inference of hand movements from local field potentials in monkey motor cortex.

The spiking of neuronal populations in motor cortex provides accurate information about movement parameters. Here we show that hand movement target and velocity can be inferred from multiple local field potentials (LFPs) in single trials approximately as efficiently as from multiple single-unit activity (SUA) recorded from the same electrodes. Our results indicate that LFPs can be used as an additional signal for decoding brain activity, particularly for new neuroprosthetic applications.

Learning-induced improvement in encoding and decoding of specific movement directions by neurons in the primary motor cortex.

Many recent studies describe learning-related changes in sensory and motor areas, but few have directly probed for improvement in neuronal coding after learning. We used information theory to analyze single-cell activity from the primary motor cortex of monkeys, before and after learning a local rotational visuomotor task. We show that after learning, neurons in the primary motor cortex conveyed more information about the direction of movement and did so with relation to their directional sensitivity. Similar to recent findings in sensory systems, this specific improvement in encoding is correlated with an increase in the slope of the neurons' tuning curve. We further demonstrate that the improved information after learning enables a more accurate reconstruction of movement direction from neuronal populations. Our results suggest that similar mechanisms govern learning in sensory and motor areas and provide further evidence for a tight relationship between the locality of learning and the properties of neurons; namely, cells only show plasticity if their preferred direction is near the training one. The results also suggest that simple learning tasks can enhance the performance of brain-machine interfaces.

Emerging patterns of neuronal responses in supplementary and primary motor areas during sensorimotor adaptation.

Acquisition and retention of sensorimotor skills have been extensively investigated psychophysically, but little is known about the underlying neuronal mechanisms. Here we examine the evolution of neural activity associated with adaptation to new kinematic tasks in two cortical areas: the caudal supplementary motor area (SMA proper), and the primary motor cortex (MI). We investigate the hypothesis that adaptation starts at premotor areas, i.e., higher in the hierarchy of computation, until a stable representation is formed in primary areas. In accordance with previous studies, we found that adaptation can be characterized by two phases: an early phase that is accompanied by fast and substantial reduction of errors, followed by a late phase with slower and more moderate improvements in behavior. We used unsupervised clustering to separate the activity of the single cells into groups of cells with similar response patterns, under the assumption that each such subpopulation forms a functional unit. We specifically observed the number of clusters in each cortical area during early and late phases of the adaptation and found that the number of clusters is higher in the SMA during early phases of adaptation. In contrast, a higher number of clusters was observed in MI only during late phases. Our results suggest a new approach to analyze responses of large populations of neurons and use it to show a hierarchy of dynamic reorganization of functional groups during adaptation.