Volitional control of single cortical neurons in a brain-machine interface

Rhesus monkeys learn to control a directional-key inspired brain machine interface via bio-feedback

Brain machine interfaces (BMI) connect brains directly to the outside world, bypassing natural neural systems and actuators. Neuronal-activity-to-motion transformation algorithms allow applications such as control of prosthetics or computer cursors. These algorithms lie within a spectrum between bio-mimetic control and bio-feedback control. The bio-mimetic approach relies on increasingly complex algorithms to decode neural activity by mimicking the natural neural system and actuator relationship while focusing on machine learning: the supervised fitting of decoder parameters. On the other hand, the bio-feedback approach uses simple algorithms and relies primarily on user learning, which may take some time, but can facilitate control of novel, non-biological appendages. An increasing amount of work has focused on the arguably more successful bio-mimetic approach. However, as chronic recordings have become more accessible and utilization of novel appendages such as computer cursors have become more universal, users can more easily spend time learning in a bio-feedback control paradigm. We believe a simple approach which leverages user learning and few assumptions will provide users with good control ability. To test the feasibility of this idea, we implemented a simple firing-rate-to-motion correspondence rule, assigned groups of neurons to virtual "directional keys" for control of a 2D cursor. Though not strictly required, to facilitate initial control, we selected neurons with similar preferred directions for each group. The groups of neurons were kept the same across multiple recording sessions to allow learning. Two Rhesus monkeys used this BMI to perform a center-out cursor movement task. After about a week of training, monkeys performed the task better and neuronal signal patterns changed on a group basis, indicating learning. While our experiments did not compare this bio-feedback BMI to bio-mimetic BMIs, the results demonstrate the feasibility of our control paradigm and paves the way for further research in multi-dimensional bio-feedback BMIs.

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.

Neural representation and modulation of volitional motivation in response to escalating efforts

Task-dependent volitional control of the selected neural activity in the cortex is critical to neuroprosthetic learning to achieve reliable and robust control of the external device. The volitional control of neural activity is driven by a motivational factor (volitional motivation), which directly reinforces the target neurons via real-time biofeedback. However, in the absence of motor behaviour, how do we evaluate volitional motivation? Here, we defined the criterion (ΔF/F) of the calcium fluorescence signal in a volitionally controlled neural task, then escalated the efforts by progressively increasing the number of reaching the criterion or holding time after reaching the criterion. We devised calcium-based progressive threshold-crossing events (termed 'Calcium PTE') and calcium-based progressive threshold-crossing holding-time (termed 'Calcium PTH') for quantitative assessment of volitional motivation in response to progressively escalating efforts. Furthermore, we used this novel neural representation of volitional motivation to explore the neural circuit and neuromodulator bases for volitional motivation. As with behavioural motivation, chemogenetic activation and pharmacological blockade of the striatopallidal pathway decreased and increased, respectively, the breakpoints of the 'Calcium PTE' and 'Calcium PTH' in response to escalating efforts. Furthermore, volitional and behavioural motivation shared similar dopamine dynamics in the nucleus accumbens in response to trial-by-trial escalating efforts. In general, the development of a neural representation of volitional motivation may open a new avenue for smooth and effective control of brain-machine interface tasks. KEY POINTS: Volitional motivation is quantitatively evaluated by M1 neural activity in response to progressively escalating volitional efforts. The striatopallidal pathway and adenosine A_2A receptor modulate volitional motivation in response to escalating efforts. Dopamine dynamics encode prediction signal for reward in response to repeated escalating efforts during motor and volitional conditioning. Mice learn to modulate neural activity to compensate for repeated escalating efforts in volitional control.

Recruitment and Differential Firing Patterns of Single Units During Conditioning to a Tone in a Mute Locked-In Human

Single units that are not related to the desired task can become related to the task by conditioning their firing rates. We theorized that, during conditioning of firing rates to a tone, (a) unrelated single units would be recruited to the task; (b) the recruitment would depend on the phase of the task; (c) tones of different frequencies would produce different patterns of single unit recruitment. In our mute locked-in participant, we conditioned single units using tones of different frequencies emitted from a tone generator. The conditioning task had three phases: Listen to the tone for 20 s, then silently sing the tone for 10 s, with a prior control period of resting for 10 s. Twenty single units were recorded simultaneously while feedback of one of the twenty single units was made audible to the mute locked-in participant. The results indicate that (a) some of the non-audible single units were recruited during conditioning, (b) some were recruited differentially depending on the phase of the paradigm (listen, rest, or silent sing), and (c) single unit firing patterns were specific for different tone frequencies such that the tone could be recognized from the pattern of single unit firings. These data are important when conditioning single unit firings in brain-computer interfacing tasks because they provide evidence that increased numbers of previously unrelated single units can be incorporated into the task. This incorporation expands the bandwidth of the recorded single unit population and thus enhances the brain-computer interface. This is the first report of conditioning of single unit firings in a human participant with a brain to computer implant.

Volitional control of individual neurons in the human brain

Brain-machine interfaces allow neuroscientists to causally link specific neural activity patterns to a particular behaviour. Thus, in addition to their current clinical applications, brain-machine interfaces can also be used as a tool to investigate neural mechanisms of learning and plasticity in the brain. Decades of research using such brain-machine interfaces have shown that animals (non-human primates and rodents) can be operantly conditioned to self-regulate neural activity in various motor-related structures of the brain. Here, we ask whether the human brain, a complex interconnected structure of over 80 billion neurons, can learn to control itself at the most elemental scale-a single neuron. We used the unique opportunity to record single units in 11 individuals with epilepsy to explore whether the firing rate of a single (direct) neuron in limbic and other memory-related brain structures can be brought under volitional control. To do this, we developed a visual neurofeedback task in which participants were trained to move a block on a screen by modulating the activity of an arbitrarily selected neuron from their brain. Remarkably, participants were able to volitionally modulate the firing rate of the direct neuron in these previously uninvestigated structures. We found that a subset of participants (learners), were able to improve their performance within a single training session. Successful learning was characterized by (i) highly specific modulation of the direct neuron (demonstrated by significantly increased firing rates and burst frequency); (ii) a simultaneous decorrelation of the activity of the direct neuron from the neighbouring neurons; and (iii) robust phase-locking of the direct neuron to local alpha/beta-frequency oscillations, which may provide some insights in to the potential neural mechanisms that facilitate this type of learning. Volitional control of neuronal activity in mnemonic structures may provide new ways of probing the function and plasticity of human memory without exogenous stimulation. Furthermore, self-regulation of neural activity in these brain regions may provide an avenue for the development of novel neuroprosthetics for the treatment of neurological conditions that are commonly associated with pathological activity in these brain structures, such as medically refractory epilepsy.

Neuron, control thyself!

This scientific commentary refers to ‘Volitional control of individual neurons in the human brain’ by Patel et al. (doi:10.1093/brain/awab370).

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.

Bypassing stroke-damaged neural pathways via a neural interface induces targeted cortical adaptation

Regaining the function of an impaired limb is highly desirable in paralyzed individuals. One possible avenue to achieve this goal is to bridge the interrupted pathway between preserved neural structures and muscles using a brain–computer interface. Here, we demonstrate that monkeys with subcortical stroke were able to learn to use an artificial cortico-muscular connection (ACMC), which transforms cortical activity into electrical stimulation to the hand muscles, to regain volitional control of a paralysed hand. The ACMC induced an adaptive change of cortical activities throughout an extensive cortical area. In a targeted manner, modulating high-gamma activity became localized around an arbitrarily-selected cortical site controlling stimulation to the muscles. This adaptive change could be reset and localized rapidly to a new cortical site. Thus, the ACMC imparts new function for muscle control to connected cortical sites and triggers cortical adaptation to regain impaired motor function after stroke.

Monkeys were trained to use an artificial cortico-muscular connection (ACMC) to regain control over a paralyzed hand following subcortical stroke. Control over the paralyzed hand was accompanied by the appearance of localized high-gamma modulation in the cortex, which could be rapidly reset and relocalized to a different cortical site to reactivate motor control.

SpikeDeeptector: a deep-learning based method for detection of neural spiking activity

OBJECTIVE

In electrophysiology, microelectrodes are the primary source for recording neural data (single unit activity). These microelectrodes can be implanted individually or in the form of arrays containing dozens to hundreds of channels. Recordings of some channels contain neural activity, which are often contaminated with noise. Another fraction of channels does not record any neural data, but only noise. By noise, we mean physiological activities unrelated to spiking, including technical artifacts and neural activities of neurons that are too far away from the electrode to be usefully processed. For further analysis, an automatic identification and continuous tracking of channels containing neural data is of great significance for many applications, e.g. automated selection of neural channels during online and offline spike sorting. Automated spike detection and sorting is also critical for online decoding in brain-computer interface (BCI) applications, in which only simple threshold crossing events are often considered for feature extraction. To our knowledge, there is no method that can universally and automatically identify channels containing neural data. In this study, we aim to identify and track channels containing neural data from implanted electrodes, automatically and more importantly universally. By universally, we mean across different recording technologies, different subjects and different brain areas.

APPROACH

We propose a novel algorithm based on a new way of feature vector extraction and a deep learning method, which we call SpikeDeeptector. SpikeDeeptector considers a batch of waveforms to construct a single feature vector and enables contextual learning. The feature vectors are then fed to a deep learning method, which learns contextualized, temporal and spatial patterns, and classifies them as channels containing neural spike data or only noise.

MAIN RESULTS

We trained the model of SpikeDeeptector on data recorded from a single tetraplegic patient with two Utah arrays implanted in different areas of the brain. The trained model was then evaluated on data collected from six epileptic patients implanted with depth electrodes, unseen data from the tetraplegic patient and data from another tetraplegic patient implanted with two Utah arrays. The cumulative evaluation accuracy was 97.20% on 1.56 million hand labeled test inputs.

SIGNIFICANCE

The results demonstrate that SpikeDeeptector generalizes not only to the new data, but also to different brain areas, subjects, and electrode types not used for training.

CLINICAL TRIAL REGISTRATION NUMBER

The clinical trial registration number for patients implanted with the Utah array is NCT01849822. For the epilepsy patients, approval from the local ethics committee at the Ruhr-University Bochum, Germany, was obtained prior to implantation.

Demonstrating Advantages of Neuromorphic Computation: A Pilot Study

Neuromorphic devices represent an attempt to mimic aspects of the brain's architecture and dynamics with the aim of replicating its hallmark functional capabilities in terms of computational power, robust learning and energy efficiency. We employ a single-chip prototype of the BrainScaleS 2 neuromorphic system to implement a proof-of-concept demonstration of reward-modulated spike-timing-dependent plasticity in a spiking network that learns to play a simplified version of the Pong video game by smooth pursuit. This system combines an electronic mixed-signal substrate for emulating neuron and synapse dynamics with an embedded digital processor for on-chip learning, which in this work also serves to simulate the virtual environment and learning agent. The analog emulation of neuronal membrane dynamics enables a 1000-fold acceleration with respect to biological real-time, with the entire chip operating on a power budget of 57 mW. Compared to an equivalent simulation using state-of-the-art software, the on-chip emulation is at least one order of magnitude faster and three orders of magnitude more energy-efficient. We demonstrate how on-chip learning can mitigate the effects of fixed-pattern noise, which is unavoidable in analog substrates, while making use of temporal variability for action exploration. Learning compensates imperfections of the physical substrate, as manifested in neuronal parameter variability, by adapting synaptic weights to match respective excitability of individual neurons.

Volitional control of single-electrode high gamma local field potentials by people with paralysis

Intracortical brain-computer interfaces (BCIs) can enable individuals to control effectors, such as a computer cursor, by directly decoding the user's movement intentions from action potentials and local field potentials (LFPs) recorded within the motor cortex. However, the accuracy and complexity of effector control achieved with such "biomimetic" BCIs will depend on the degree to which the intended movements used to elicit control modulate the neural activity. In particular, channels that do not record distinguishable action potentials and only record LFP modulations may be of limited use for BCI control. In contrast, a biofeedback approach may surpass these limitations by letting the participants generate new control signals and learn strategies that improve the volitional control of signals used for effector control. Here, we show that, by using a biofeedback paradigm, three individuals with tetraplegia achieved volitional control of gamma LFPs (40-400 Hz) recorded by a single microelectrode implanted in the precentral gyrus. Control was improved over a pair of consecutive sessions up to 3 days apart. In all but one session, the channel used to achieve control lacked distinguishable action potentials. Our results indicate that biofeedback LFP-based BCIs may potentially contribute to the neural modulation necessary to obtain reliable and useful control of effectors. NEW & NOTEWORTHY Our study demonstrates that people with tetraplegia can volitionally control individual high-gamma local-field potential (LFP) channels recorded from the motor cortex, and that this control can be improved using biofeedback. Motor cortical LFP signals are thought to be both informative and stable intracortical signals and, thus, of importance for future brain-computer interfaces.

Invasive Brain Machine Interface System

Because of high spatial-temporal resolution of neural signals obtained by invasive recording, the invasive brain-machine interfaces (BMI) have achieved great progress in the past two decades. With success in animal research, BMI technology is transferring to clinical trials for helping paralyzed people to restore their lost motor functions. This chapter gives a brief review of BMI development from animal experiments to human clinical studies in the following aspects: (1) BMIs based on rodent animals; (2) BMI based on non-human primates; and (3) pilot BMIs studies in clinical trials. In the end, the chapter concludes with a summary of potential opportunities and future challenges in BMI technology.

Optimizing the Usability of Brain-Computer Interfaces

Brain-computer interfaces are in the process of moving from the laboratory to the clinic. These devices act by reading neural activity and using it to directly control a device, such as a cursor on a computer screen. An open question in the field is how to map neural activity to device movement in order to achieve the most proficient control. This question is complicated by the fact that learning, especially the long-term skill learning that accompanies weeks of practice, can allow subjects to improve performance over time. Typical approaches to this problem attempt to maximize the biomimetic properties of the device in order to limit the need for extensive training. However, it is unclear if this approach would ultimately be superior to performance that might be achieved with a nonbiomimetic device once the subject has engaged in extended practice and learned how to use it. Here we approach this problem using ideas from optimal control theory. Under the assumption that the brain acts as an optimal controller, we present a formal definition of the usability of a device and show that the optimal postlearning mapping can be written as the solution of a constrained optimization problem. We then derive the optimal mappings for particular cases common to most brain-computer interfaces. Our results suggest that the common approach of creating biomimetic interfaces may not be optimal when learning is taken into account. More broadly, our method provides a blueprint for optimal device design in general control-theoretic contexts.

Open-Source, Low Cost, Free-Behavior Monitoring, and Reward System for Neuroscience Research in Non-human Primates

We describe a low-cost system designed to document bodily movement and neural activity and deliver rewards to monkeys behaving freely in their home cage. An important application is to studying brain-machine interface (BMI) systems during free behavior, since brain signals associated with natural movement can differ significantly from those associated with more commonly used constrained conditions. Our approach allows for short-latency (<500 ms) reward delivery and behavior monitoring using low-cost off-the-shelf components. This system interfaces existing untethered recording equipment with a custom hub that controls a cage-mounted feeder. The behavior monitoring system uses a depth camera to provide real-time, easy-to-analyze, gross movement data streams. In a proof-of-concept experiment we demonstrate robust learning of neural activity using the system over 14 behavioral sessions.

Decoding Local Field Potentials for Neural Interfaces

The stability and frequency content of local field potentials (LFPs) offer key advantages for long-term, low-power neural interfaces. However, interpreting LFPs may require new signal processing techniques which should be informed by a scientific understanding of how these recordings arise from the coordinated activity of underlying neuronal populations. We review current approaches to decoding LFPs for brain-machine interface (BMI) applications, and suggest several directions for future research. To facilitate an improved understanding of the relationship between LFPs and spike activity, we share a dataset of multielectrode recordings from monkey motor cortex, and describe two unsupervised analysis methods we have explored for extracting a low-dimensional feature space that is amenable to biomimetic decoding and biofeedback training.

Spatial co-adaptation of cortical control columns in a micro-ECoG brain-computer interface

OBJECTIVE

Electrocorticography (ECoG) has been used for a range of applications including electrophysiological mapping, epilepsy monitoring, and more recently as a recording modality for brain-computer interfaces (BCIs). Studies that examine ECoG electrodes designed and implanted chronically solely for BCI applications remain limited. The present study explored how two key factors influence chronic, closed-loop ECoG BCI: (i) the effect of inter-electrode distance on BCI performance and (ii) the differences in neural adaptation and performance when fixed versus adaptive BCI decoding weights are used.

APPROACH

The amplitudes of epidural micro-ECoG signals between 75 and 105 Hz with 300 μm diameter electrodes were used for one-dimensional and two-dimensional BCI tasks. The effect of inter-electrode distance on BCI control was tested between 3 and 15 mm. Additionally, the performance and cortical modulation differences between constant, fixed decoding using a small subset of channels versus adaptive decoding weights using the entire array were explored.

MAIN RESULTS

Successful BCI control was possible with two electrodes separated by 9 and 15 mm. Performance decreased and the signals became more correlated when the electrodes were only 3 mm apart. BCI performance in a 2D BCI task improved significantly when using adaptive decoding weights (80%-90%) compared to using constant, fixed weights (50%-60%). Additionally, modulation increased for channels previously unavailable for BCI control under the fixed decoding scheme upon switching to the adaptive, all-channel scheme.

SIGNIFICANCE

Our results clearly show that neural activity under a BCI recording electrode (which we define as a 'cortical control column') readily adapts to generate an appropriate control signal. These results show that the practical minimal spatial resolution of these control columns with micro-ECoG BCI is likely on the order of 3 mm. Additionally, they show that the combination and interaction between neural adaptation and machine learning are critical to optimizing ECoG BCI performance.

Empirical Movement Models for Brain Computer Interfaces

For brain-computer interfaces (BCIs) which provide the user continuous position control, there is little standardization of performance metrics or evaluative tasks. One candidate metric is Fitts's law, which has been used to describe aimed movements across a range of computer interfaces, and has recently been applied to BCI tasks. Reviewing selected studies, we identify two basic problems with Fitts's law: its predictive performance is fragile, and the estimation of 'information transfer rate' from the model is unsupported. Our main contribution is the adaptation and validation of an alternative model to Fitts's law in the BCI context. We show that the Shannon-Welford model outperforms Fitts's law, showing robust predictive power when target distance and width have disproportionate effects on difficulty. Building on a prior study of the Shannon-Welford model, we show that identified model parameters offer a novel approach to quantitatively assess the role of control-display gain in speed/accuracy performance tradeoffs during brain control.

Intracortical Brain-Machine Interfaces Advance Sensorimotor Neuroscience

Brain-machine interfaces (BMIs) decode brain activity to control external devices. Over the past two decades, the BMI community has grown tremendously and reached some impressive milestones, including the first human clinical trials using chronically implanted intracortical electrodes. It has also contributed experimental paradigms and important findings to basic neuroscience. In this review, we discuss neuroscience achievements stemming from BMI research, specifically that based upon upper limb prosthetic control with intracortical microelectrodes. We will focus on three main areas: first, we discuss progress in neural coding of reaches in motor cortex, describing recent results linking high dimensional representations of cortical activity to muscle activation. Next, we describe recent findings on learning and plasticity in motor cortex on various time scales. Finally, we discuss how bidirectional BMIs have led to better understanding of somatosensation in and related to motor cortex.

Brain-computer interfaces for dissecting cognitive processes underlying sensorimotor control

Sensorimotor control engages cognitive processes such as prediction, learning, and multisensory integration. Understanding the neural mechanisms underlying these cognitive processes with arm reaching is challenging because we currently record only a fraction of the relevant neurons, the arm has nonlinear dynamics, and multiple modalities of sensory feedback contribute to control. A brain-computer interface (BCI) is a well-defined sensorimotor loop with key simplifying advantages that address each of these challenges, while engaging similar cognitive processes. As a result, BCI is becoming recognized as a powerful tool for basic scientific studies of sensorimotor control. Here, we describe the benefits of BCI for basic scientific inquiries and review recent BCI studies that have uncovered new insights into the neural mechanisms underlying sensorimotor control.

A cortical-spinal prosthesis for targeted limb movement in paralysed primate avatars

Motor paralysis is among the most disabling aspects of injury to the central nervous system. Here we develop and test a target-based cortical-spinal neural prosthesis that employs neural activity recorded from pre-motor neurons to control limb movements in functionally paralyzed primate avatars. Given the complexity by which muscle contractions are naturally controlled, we approach the problem of eliciting goal-directed limb movement in paralyzed animals by focusing on the intended targets of movement rather than their intermediate trajectories. We then match this information in real-time with spinal cord and muscle stimulation parameters that produce free planar limb movements to those intended target locations. We demonstrate that both the decoded activities of pre-motor populations and their adaptive responses can be used, after brief training, to effectively direct an avatar’s limb to distinct targets variably displayed on a screen. These findings advance the future possibility of reconstituting targeted limb movement in paralyzed subjects.

Simultaneous and independent control of a brain-computer interface and contralateral limb movement

Toward expanding the population of potential BCI users to the many individuals with lateralized cortical stroke, here we examined whether the cortical hemisphere controlling ongoing movements of the contralateral limb can simultaneously generate signals to control a BCI. A monkey was trained to perform a simultaneous BCI and manual control task designed to test whether one hemisphere could effectively differentiate its output and provide independent control of two tasks. Pairs of well-isolated single units were used to control a BCI cursor in one dimension, while isometric wrist torque of the contralateral forelimb controlled the cursor in a second dimension. The monkey could independently modulate cortical units and contralateral wrist torque regardless of the strength of directional tuning of the units controlling the BCI. When the presented targets required explicit decoupling of unit activity and wrist torque, directionally tuned units exhibited significantly less efficient cursor trajectories compared to when unit activity and wrist torque could remain correlated. The results indicate that neural activity from a single hemisphere can be effectively decoupled to simultaneously control a BCI and ongoing limb movement, suggesting that BCIs may be a viable future treatment for individuals with lateralized cortical stroke.

Sequential activation of premotor, primary somatosensory and primary motor areas in humans during cued finger movements

OBJECTIVE

Human voluntary movements are a final product of complex interactions between multiple sensory, cognitive and motor areas of central nervous system. The objective was to investigate temporal sequence of activation of premotor (PM), primary motor (M1) and somatosensory (S1) areas during cued finger movements.

METHODS

Electrocorticography (ECoG) was used to measure activation timing in human PM, S1, and M1 neurons in preparation for finger movements in 5 subjects with subdural grids for seizure localization. Cortical activation was determined by the onset of high gamma (HG) oscillation (70-150Hz). The three cortical regions were mapped anatomically using a common brain atlas and confirmed independently with direct electrical cortical stimulation, somatosensory evoked potentials and detection of HG response to tactile stimulation. Subjects were given visual cues to flex each finger or pinch the thumb and index finger. Movements were captured with a dataglove and time-locked with ECoG. A windowed covariance metric was used to identify the rising slope of HG power between two electrodes and compute time lag. Statistical constraints were applied to the time estimates to combat the noise. Rank sum testing was used to verify the sequential activation of cortical regions across 5 subjects.

RESULTS

In all 5 subjects, HG activation in PM preceded S1 by an average of 53±13ms (P=0.03), PM preceded M1 by 180±40ms (P=0.001) and S1 activation preceded M1 by 136±40ms (P=0.04).

CONCLUSIONS

Sequential HG activation of PM, S1 and M1 regions in preparation for movements is reported. Activity in S1 prior to any overt body movements supports the notion that these neurons may encode sensory information in anticipation of movements, i.e., an efference copy. Our analysis suggests that S1 modulation likely originates from PM.

SIGNIFICANCE

First electrophysiological evidence of efference copy in humans.

Intracortical recording interfaces: current challenges to chronic recording function

Brain Computer Interfaces (BCIs) offer significant hope to tetraplegic and paraplegic individuals. This technology relies on extracting and translating motor intent to facilitate control of a computer cursor or to enable fine control of an external assistive device such as a prosthetic limb. Intracortical recording interfaces (IRIs) are critical components of BCIs and consist of arrays of penetrating electrodes that are implanted into the motor cortex of the brain. These multielectrode arrays (MEAs) are responsible for recording and conducting neural signals from local ensembles of neurons in the motor cortex with the high speed and spatiotemporal resolution that is required for exercising control of external assistive prostheses. Recent design and technological innovations in the field have led to significant improvements in BCI function. However, long-term (chronic) BCI function is severely compromised by short-term (acute) IRI recording failure. In this review, we will discuss the design and function of current IRIs. We will also review a host of recent advances that contribute significantly to our overall understanding of the cellular and molecular events that lead to acute recording failure of these invasive implants. We will also present recent improvements to IRI design and provide insights into the futuristic design of more chronically functional IRIs.

The current state of the brain-computer interface problem

It was only 40 years ago that the first PC appeared. Over this period, rather short in historical terms, we have witnessed the revolutionary changes in lives of individuals and the entire society. Computer technologies are tightly connected with any field, either directly or indirectly. We can currently claim that computers are manifold superior to a human mind in terms of a number of parameters; however, machines lack the key feature: they are incapable of independent thinking (like a human). However, the key to successful development of humankind is collaboration between the brain and the computer rather than competition. Such collaboration when a computer broadens, supplements, or replaces some brain functions is known as the brain-computer interface. Our review focuses on real-life implementation of this collaboration.

Brain-machine interface for eye movements

A number of studies in tetraplegic humans and healthy nonhuman primates (NHPs) have shown that neuronal activity from reach-related cortical areas can be used to predict reach intentions using brain-machine interfaces (BMIs) and therefore assist tetraplegic patients by controlling external devices (e.g., robotic limbs and computer cursors). However, to our knowledge, there have been no studies that have applied BMIs to eye movement areas to decode intended eye movements. In this study, we recorded the activity from populations of neurons from the lateral intraparietal area (LIP), a cortical node in the NHP saccade system. Eye movement plans were predicted in real time using Bayesian inference from small ensembles of LIP neurons without the animal making an eye movement. Learning, defined as an increase in the prediction accuracy, occurred at the level of neuronal ensembles, particularly for difficult predictions. Population learning had two components: an update of the parameters of the BMI based on its history and a change in the responses of individual neurons. These results provide strong evidence that the responses of neuronal ensembles can be shaped with respect to a cost function, here the prediction accuracy of the BMI. Furthermore, eye movement plans could be decoded without the animals emitting any actual eye movements and could be used to control the position of a cursor on a computer screen. These findings show that BMIs for eye movements are promising aids for assisting paralyzed patients.

Synapse rearrangements upon learning: from divergent-sparse connectivity to dedicated sub-circuits

Learning can involve formation of new synapses and loss of synapses, providing memory traces of learned skills. Recent findings suggest that these synapse rearrangements reflect assembly of task-related sub-circuits from initially broadly distributed and sparse connectivity in the brain. These local circuit remodeling processes involve rapid emergence of synapses upon learning, followed by protracted validation involving strengthening of some new synapses, and selective elimination of others. The timing of these consolidation processes can vary. Here, we review these findings, focusing on how molecular/cellular mechanisms of synapse assembly, strengthening, and elimination might interface with circuit/system mechanisms of learning and memory consolidation. An integrated understanding of these learning-related processes should provide a better basis to elucidate how experience, genetic background, and disease influence brain function.

Delineating the whole brain BOLD response to passive movement kinematics

The field of brain-machine interfaces (BMIs) has made great advances in recent years, converting thought to movement, with some of the most successful implementations measuring directly from the motor cortex. However, the ability to record from additional regions of the brain could potentially improve flexibility and robustness of use. In addition, BMIs of the future will benefit from integrating kinesthesia into the control loop. Here, we examine whether changes in passively induced forefinger movement amplitude are represented in different regions than forefinger velocity via a MR compatible robotic manipulandum. Using functional magnetic resonance imaging (fMRI), five healthy participants were exposed to combinations of forefinger movement amplitude and velocity in a factorial design followed by an epoch-based analysis. We found that primary and secondary somatosensory regions were activated, as well as cingulate motor area, putamen and cerebellum, with greater activity from changes in velocity compared to changes in amplitude. This represents the first investigation into whole brain response to parametric changes in passive movement kinematics. In addition to informing BMIs, these results have implications towards neural correlates of robotic rehabilitation.

A brain-machine-muscle interface for restoring hindlimb locomotion after complete spinal transection in rats

A brain-machine interface (BMI) is a neuroprosthetic device that can restore motor function of individuals with paralysis. Although the feasibility of BMI control of upper-limb neuroprostheses has been demonstrated, a BMI for the restoration of lower-limb motor functions has not yet been developed. The objective of this study was to determine if gait-related information can be captured from neural activity recorded from the primary motor cortex of rats, and if this neural information can be used to stimulate paralysed hindlimb muscles after complete spinal cord transection. Neural activity was recorded from the hindlimb area of the primary motor cortex of six female Sprague Dawley rats during treadmill locomotion before and after mid-thoracic transection. Before spinal transection there was a strong association between neural activity and the step cycle. This association decreased after spinal transection. However, the locomotive state (standing vs. walking) could still be successfully decoded from neural recordings made after spinal transection. A novel BMI device was developed that processed this neural information in real-time and used it to control electrical stimulation of paralysed hindlimb muscles. This system was able to elicit hindlimb muscle contractions that mimicked forelimb stepping. We propose this lower-limb BMI as a future neuroprosthesis for human paraplegics.

Bidirectional control of a one-dimensional robotic actuator by operant conditioning of a single unit in rat motor cortex

The design of efficient neuroprosthetic devices has become a major challenge for the long-term goal of restoring autonomy to motor-impaired patients. One approach for brain control of actuators consists in decoding the activity pattern obtained by simultaneously recording large neuronal ensembles in order to predict in real-time the subject's intention, and move the prosthesis accordingly. An alternative way is to assign the output of one or a few neurons by operant conditioning to control the prosthesis with rules defined by the experimenter, and rely on the functional adaptation of these neurons during learning to reach the desired behavioral outcome. Here, several motor cortex neurons were recorded simultaneously in head-fixed awake rats and were conditioned, one at a time, to modulate their firing rate up and down in order to control the speed and direction of a one-dimensional actuator carrying a water bottle. The goal was to maintain the bottle in front of the rat's mouth, allowing it to drink. After learning, all conditioned neurons modulated their firing rate, effectively controlling the bottle position so that the drinking time was increased relative to chance. The mean firing rate averaged over all bottle trajectories depended non-linearly on position, so that the mouth position operated as an attractor. Some modifications of mean firing rate were observed in the surrounding neurons, but to a lesser extent. Notably, the conditioned neuron reacted faster and led to a better control than surrounding neurons, as calculated by using the activity of those neurons to generate simulated bottle trajectories. Our study demonstrates the feasibility, even in the rodent, of using a motor cortex neuron to control a prosthesis in real-time bidirectionally. The learning process includes modifications of the activity of neighboring cortical neurons, while the conditioned neuron selectively leads the activity patterns associated with the prosthesis control.

Restoring sensorimotor function through intracortical interfaces: progress and looming challenges

The loss of a limb or paralysis resulting from spinal cord injury has devastating consequences on quality of life. One approach to restoring lost sensory and motor abilities in amputees and patients with tetraplegia is to supply them with implants that provide a direct interface with the CNS. Such brain-machine interfaces might enable a patient to exert voluntary control over a prosthetic or robotic limb or over the electrically induced contractions of paralysed muscles. A parallel interface could convey sensory information about the consequences of these movements back to the patient. Recent developments in the algorithms that decode motor intention from neuronal activity and in approaches to convey sensory feedback by electrically stimulating neurons, using biomimetic and adaptation-based approaches, have shown the promise of invasive interfaces with sensorimotor cortices, although substantial challenges remain.

What limits the performance of current invasive brain machine interfaces?

The concept of a brain-machine interface (BMI) or a computer-brain interface is simple: BMI creates a communication pathway for a direct control by brain of an external device. In reality BMIs are very complex devices and only recently the increase in computing power of microprocessors enabled a boom in BMI research that continues almost unabated to this date, the high point being the insertion of electrode arrays into the brains of 5 human patients in a clinical trial run by Cyberkinetics with few other clinical tests still in progress. Meanwhile several EEG-based BMI devices (non-invasive BMIs) were launched commercially. Modern electronics and dry electrode technology made possible to drive the cost of some of these devices below few hundred dollars. However, the initial excitement of the direct control by brain waves of a computer or other equipment is dampened by large efforts required for learning, high error rates and slow response speed. All these problems are directly related to low information transfer rates typical for such EEG-based BMIs. In invasive BMIs employing multiple electrodes inserted into the brain one may expect much higher information transfer rates than in EEG-based BMIs because, in theory, each electrode provides an independent information channel. However, although invasive BMIs require more expensive equipment and have ethical problems related to the need to insert electrodes in the live brain, such financial and ethical costs are often not offset by a dramatic improvement in the information transfer rate. Thus the main topic of this review is why in invasive BMIs an apparently much larger information content obtained with multiple extracellular electrodes does not translate into much higher rates of information transfer? This paper explores possible answers to this question by concluding that more research on what movement parameters are encoded by neurons in motor cortex is needed before we can enjoy the next generation BMIs.

A reductionist approach to the analysis of learning in brain-computer interfaces

The complexity and scale of brain-computer interface (BCI) studies limit our ability to investigate how humans learn to use BCI systems. It also limits our capacity to develop adaptive algorithms needed to assist users with their control. Adaptive algorithm development is forced offline and typically uses static data sets. But this is a poor substitute for the online, dynamic environment where algorithms are ultimately deployed and interact with an adapting user. This work evaluates a paradigm that simulates the control problem faced by human subjects when controlling a BCI, but which avoids the many complications associated with full-scale BCI studies. Biological learners can be studied in a reductionist way as they solve BCI-like control problems, and machine learning algorithms can be developed and tested in closed loop with the subjects before being translated to full BCIs. The method is to map 19 joint angles of the hand (representing neural signals) to the position of a 2D cursor which must be piloted to displayed targets (a typical BCI task). An investigation is presented on how closely the joint angle method emulates BCI systems; a novel learning algorithm is evaluated, and a performance difference between genders is discussed.

Affective Brain-Computer Interfaces As Enabling Technology for Responsive Psychiatric Stimulation

There is a pressing clinical need for responsive neurostimulators, which sense a patient's brain activity and deliver targeted electrical stimulation to suppress unwanted symptoms. This is particularly true in psychiatric illness, where symptoms can fluctuate throughout the day. Affective BCIs, which decode emotional experience from neural activity, are a candidate control signal for responsive stimulators targeting the limbic circuit. Present affective decoders, however, cannot yet distinguish pathologic from healthy emotional extremes. Indiscriminate stimulus delivery would reduce quality of life and may be actively harmful. We argue that the key to overcoming this limitation is to specifically decode volition, in particular the patient's intention to experience emotional regulation. Those emotion-regulation signals already exist in prefrontal cortex (PFC), and could be extracted with relatively simple BCI algorithms. We describe preliminary data from an animal model of PFC-controlled limbic brain stimulation and discuss next steps for pre-clinical testing and possible translation.

Creating new functional circuits for action via brain-machine interfaces

Brain-machine interfaces (BMIs) are an emerging technology with great promise for developing restorative therapies for those with disabilities. BMIs also create novel, well-defined functional circuits for action that are distinct from the natural sensorimotor apparatus. Closed-loop control of BMI systems can also actively engage learning and adaptation. These properties make BMIs uniquely suited to study learning of motor and non-physical, abstract skills. Recent work used motor BMIs to shed light on the neural representations of skill formation and motor adaptation. Emerging work in sensory BMIs, and other novel interface systems, also highlight the promise of using BMI systems to study fundamental questions in learning and sensorimotor control. This paper outlines the interpretation of BMIs as novel closed-loop systems and the benefits of these systems for studying learning. We review BMI learning studies, their relation to motor control, and propose future directions for this nascent field. Understanding learning in BMIs may both elucidate mechanisms of natural motor and abstract skill learning, and aid in developing the next generation of neuroprostheses.

Restoration of upper limb movement via artificial corticospinal and musculospinal connections in a monkey with spinal cord injury

Functional loss of limb control in individuals with spinal cord injury or stroke can be caused by interruption of corticospinal pathways, although the neural circuits located above and below the lesion remain functional. An artificial neural connection that bridges the lost pathway and connects cortical to spinal circuits has potential to ameliorate the functional loss. We investigated the effects of introducing novel artificial neural connections in a paretic monkey that had a unilateral spinal cord lesion at the C2 level. The first application bridged the impaired spinal lesion. This allowed the monkey to drive the spinal stimulation through volitionally controlled power of high-gamma activity in either the premotor or motor cortex, and thereby to acquire a force-matching target. The second application created an artificial recurrent connection from a paretic agonist muscle to a spinal site, allowing muscle-controlled spinal stimulation to boost on-going activity in the muscle. These results suggest that artificial neural connections can compensate for interrupted descending pathways and promote volitional control of upper limb movement after damage of descending pathways such as spinal cord injury or stroke.

Movement representation in the primary motor cortex and its contribution to generalizable EMG predictions

It is well known that discharge of neurons in the primary motor cortex (M1) depends on end-point force and limb posture. However, the details of these relations remain unresolved. With the development of brain-machine interfaces (BMIs), these issues have taken on practical as well as theoretical importance. We examined how the M1 encodes movement by comparing single-neuron and electromyographic (EMG) preferred directions (PDs) and by predicting force and EMGs from multiple neurons recorded during an isometric wrist task. Monkeys moved a cursor from a central target to one of eight peripheral targets by exerting force about the wrist while the forearm was held in one of two postures. We fit tuning curves to both EMG and M1 activity measured during the hold period, from which we computed both PDs and the change in PD between forearm postures (ΔPD). We found a unimodal distribution of these ΔPDs, the majority of which were intermediate between the typical muscle response and an unchanging, extrinsic coordinate system. We also discovered that while most neuron-to-EMG predictions generalized well across forearm postures, end-point force measured in extrinsic coordinates did not. The lack of force generalization was due to musculoskeletal changes with posture. Our results show that the dynamics of most of the recorded M1 signals are similar to those of muscle activity and imply that a BMI designed to drive an actuator with dynamics like those of muscles might be more robust and easier to learn than a BMI that commands forces or movements in external coordinates.

Restoration of grasp following paralysis through brain-controlled stimulation of muscles

Patients with spinal cord injury lack the connections between brain and spinal cord circuits essential for voluntary movement. Clinical systems that achieve muscle contraction through functional electrical stimulation (FES) have proven to be effective in allowing patients with tetraplegia to regain control of hand movement and to achieve a greater measure of independence in activities of daily living ^1,2. In typical systems, the patient uses residual proximal limb movements to trigger pre-programmed stimulation that causes the paralyzed muscles to contract, allowing use of one or two basic grasps. Instead, we have developed, in primates, an FES system that is controlled by recordings made from microelectrodes permanently implanted in the brain. We simulated some of the effects of the paralysis caused by C5-C6 spinal cord injury ³ by injecting a local anesthetic to block the median and ulnar nerves at the elbow. Then, using recordings from approximately 100 neurons in the motor cortex, we predicted the intended activity of several of the paralyzed muscles, and used these predictions to control the intensity of stimulation of the same muscles. This process essentially bypassed the spinal cord, restoring to the monkeys voluntary control of their paralyzed muscles. This achievement represents a major advance toward similar restoration of hand function in human patients through brain-controlled FES. We anticipate that in human patients, this neuroprosthesis would allow much more flexible and dexterous use of the hand than is possible with existing FES systems.

Packaging and Non-Hermetic Encapsulation Technology for Flip Chip on Implantable MEMS Devices

We report here a successful demonstration of a flip-chip packaging approach for a microelectromechanical systems (MEMS) device with in-plane movable microelectrodes implanted in a rodent brain. The flip-chip processes were carried out using a custom-made apparatus that was capable of the following: 1) creating Ag epoxy microbumps for first-level interconnect; 2) aligning the die and the glass substrate; and 3) creating non-hermetic encapsulation (NHE). The completed flip-chip package had an assembled weight of only 0.5 g significantly less than the previously designed wire-bonded package of 4.5 g. The resistance of the Ag bumps was found to be negligible. The MEMS micro-electrodes were successfully tested for its mechanical movement with microactuators generating forces of 450 μN with a displacement resolution of 8.8 μm/step. An NHE on the front edge of the package was created by patterns of hydrophobic silicone microstructures to prevent contamination from cerebrospinal fluid while simultaneously allowing the microelectrodes to move in and out of the package boundary. The breakdown pressure of the NHE was found to be 80 cm of water, which is significantly (4.5-11 times) larger than normal human intracranial pressures. Bench top tests and in vivo tests of the MEMS flip-chip packages for up to 75 days showed reliable NHE for potential long-term implantation.

Applying best practices from digital control systems to BMI implementation

Many brain-machine interface (BMI) algorithms, such as the population vector decoder, must estimate neural spike rates before transforming this information into an external output signal. Often, rate estimation is performed via the selection of a bin width corresponding to the effective sampling rate of the decoding algorithm. Here, we implement real-time rate estimation by extending prior work on the optimization of Gaussian filters for offline rate estimation. We show that higher sampling rates result in improved spike rate estimation. We further show that the choice of sampling rate need not dictate the number of parameters which must be used in an autoregressive decoding algorithm. Multiple studies in other neural signal processing contexts suggest that BMI performance could be improved substantially via careful choice of smoothing filter, discrete-time decoder representation, and sampling rate. Together, these ensure minimal deviation from the behavior of the modeled continuous-time systems.