Using reinforcement learning to provide stable brain-machine interface control despite neural input reorganization

An in-silico framework for modeling optimal control of neural systems

Introduction

Brain-machine interfaces have reached an unprecedented capacity to measure and drive activity in the brain, allowing restoration of impaired sensory, cognitive or motor function. Classical control theory is pushed to its limit when aiming to design control laws that are suitable for large-scale, complex neural systems. This work proposes a scalable, data-driven, unified approach to study brain-machine-environment interaction using established tools from dynamical systems, optimal control theory, and deep learning.

Methods

To unify the methodology, we define the environment, neural system, and prosthesis in terms of differential equations with learnable parameters, which effectively reduce to recurrent neural networks in the discrete-time case. Drawing on tools from optimal control, we describe three ways to train the system: Direct optimization of an objective function, oracle-based learning, and reinforcement learning. These approaches are adapted to different assumptions about knowledge of system equations, linearity, differentiability, and observability.

Results

We apply the proposed framework to train an in-silico neural system to perform tasks in a linear and a nonlinear environment, namely particle stabilization and pole balancing. After training, this model is perturbed to simulate impairment of sensor and motor function. We show how a prosthetic controller can be trained to restore the behavior of the neural system under increasing levels of perturbation.

Discussion

We expect that the proposed framework will enable rapid and flexible synthesis of control algorithms for neural prostheses that reduce the need for in-vivo testing. We further highlight implications for sparse placement of prosthetic sensor and actuator components.

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.

Neural Decoders Using Reinforcement Learning in Brain Machine Interfaces: A Technical Review

Creating flexible and robust brain machine interfaces (BMIs) is currently a popular topic of research that has been explored for decades in medicine, engineering, commercial, and machine-learning communities. In particular, the use of techniques using reinforcement learning (RL) has demonstrated impressive results but is under-represented in the BMI community. To shine more light on this promising relationship, this article aims to provide an exhaustive review of RL's applications to BMIs. Our primary focus in this review is to provide a technical summary of various algorithms used in RL-based BMIs to decode neural intention, without emphasizing preprocessing techniques on the neural signals and reward modeling for RL. We first organize the literature based on the type of RL methods used for neural decoding, and then each algorithm's learning strategy is explained along with its application in BMIs. A comparative analysis highlighting the similarities and uniqueness among neural decoders is provided. Finally, we end this review with a discussion about the current stage of RLBMIs including their limitations and promising directions for future research.

Kernel Temporal Differences for EEG-based Reinforcement Learning Brain Machine Interfaces

Kernel temporal differences (KTD) (λ) algorithm integrated in Q-learning (Q-KTD) has shown its applicability and feasibility for reinforcement learning brain machine interfaces (RLBMIs). RLBMI with its unique learning strategy based on trial-error allows continuous learning and adaptation in BMIs. Q-KTD has shown good performance in both open and closed-loop experiments for finding a proper mapping from neural intention to control commands of an external device. However, previous studies have been limited to intracortical BMIs where monkey's firing rates from primary motor cortex were used as inputs to the neural decoder. This study provides the first attempt to investigate Q-KTD algorithm's applicability in EEG-based RLBMIs. Two different publicly available EEG data sets are considered, we refer to them as Data set A and Data set B. EEG motor imagery tasks are integrated in a single step center-out reaching task, and we observe the open-loop RLBMI experiments reach 100% average success rates after sufficient learning experience. Data set A converges after approximately 20 epochs for raw features and Data set B shows convergence after approximately 40 epochs for both raw and Fourier transform features. Although there still exist challenges to overcome in EEG-based RLBMI using Q-KTD, including increasing the learning speed, and optimization of a continuously growing number of kernel units, the results encourage further investigation of Q-KTD in closed-loop RLBMIs using EEG. Clinical Relevance- This study supports feasibility of noninvasive EEG-based RLBMI implementations and addresses benefits and challenges of RLBMI using EEG.

A behavioral paradigm for cortical control of a robotic actuator by freely moving rats in a one-dimensional two-target reaching task

BACKGROUND

Controlling the trajectory of a neuroprosthesis to reach distant targets is a commonly used brain-machine interface (BMI) task in primates and has not been available for rodents yet.

NEW METHOD

Here, we describe a novel, fine-tuned behavioral paradigm and setup which enables this task for rats in one-dimensional space for reaching two distant targets depending on their limited cognitive and visual capabilities compared to those of primates. An online transform was used to convert the activity of a pair of primary motor cortex (M1) units into two robotic actions. The rats were shaped to adapt to the transform and direct the robotic actuator toward the selected target by modulating the activity of the M1 neurons.

RESULTS

All three rats involved in the study were capable of achieving randomly selected targets with at least 78% accuracy. A total of 9 out of 16 pairs of units examined were eligible for exceeding this success criterion. Two out of three rats were capable of reversal learning, where the mapping between the activity of the M1 units and the robotic actions were reversed.

COMPARISON WITH EXISTING METHODS

The present work is the first demonstration of trajectory-based control of a neuroprosthetic device by rodents to reach two distant targets using visual feedback.

CONCLUSION

The behavioral paradigm and setup introduced here can be used as a cost-effective platform for elucidating the information processing principles in the neural circuits related to neuroprosthetic control and for studying the performance of novel BMI technologies using freely moving rats.

Mirror neurons are modulated by grip force and reward expectation in the sensorimotor cortices (S1, M1, PMd, PMv)

Mirror Neurons (MNs) respond similarly when primates make or observe grasping movements. Recent work indicates that reward expectation influences rostral M1 (rM1) during manual, observational, and Brain Machine Interface (BMI) reaching movements. Previous work showed MNs are modulated by subjective value. Here we expand on the above work utilizing two non-human primates (NHPs), one male Macaca Radiata (NHP S) and one female Macaca Mulatta (NHP P), that were trained to perform a cued reward level isometric grip-force task, where the NHPs had to apply visually cued grip-force to move and transport a virtual object. We found a population of (S1 area 1–2, rM1, PMd, PMv) units that significantly represented grip-force during manual and observational trials. We found the neural representation of visually cued force was similar during observational trials and manual trials for the same units; however, the representation was weaker during observational trials. Comparing changes in neural time lags between manual and observational tasks indicated that a subpopulation fit the standard MN definition of observational neural activity lagging the visual information. Neural activity in (S1 areas 1–2, rM1, PMd, PMv) significantly represented force and reward expectation. In summary, we present results indicating that sensorimotor cortices have MNs for visually cued force and value.

On the Impact of Gravity Compensation on Reinforcement Learning in Goal-Reaching Tasks for Robotic Manipulators

Advances in machine learning technologies in recent years have facilitated developments in autonomous robotic systems. Designing these autonomous systems typically requires manually specified models of the robotic system and world when using classical control-based strategies, or time consuming and computationally expensive data-driven training when using learning-based strategies. Combination of classical control and learning-based strategies may mitigate both requirements. However, the performance of the combined control system is not obvious given that there are two separate controllers. This paper focuses on one such combination, which uses gravity-compensation together with reinforcement learning (RL). We present a study of the effects of gravity compensation on the performance of two reinforcement learning algorithms when solving reaching tasks using a simulated seven-degree-of-freedom robotic arm. The results of our study demonstrate that gravity compensation coupled with RL can reduce the training required in reaching tasks involving elevated target locations, but not all target locations.

The value-complexity trade-off for reinforcement learning based brain-computer interfaces

OBJECTIVE

One of the recent developments in the field of brain-computer interfaces (BCI) is the reinforcement learning (RL) based BCI paradigm, which uses neural error responses as the reward feedback on the agent's action. While having several advantages over motor imagery based BCI, the reliability of RL-BCI is critically dependent on the decoding accuracy of noisy neural error signals. A principled method is needed to optimally handle this inherent noise under general conditions.

APPROACH

By determining a trade-off between the expected value and the informational cost of policies, the info-RL (IRL) algorithm provides optimal low-complexity policies, which are robust under noisy reward conditions and achieve the maximal obtainable value. In this work we utilize the IRL algorithm to characterize the maximal obtainable value under different noise levels, which in turn is used to extract the optimal robust policy for each noise level.

MAIN RESULTS

Our simulation results of a setting with Gaussian noise show that the complexity level of the optimal policy is dependent on the reward magnitude but not on the reward variance, whereas the variance determines whether a lower complexity solution is favorable or not. We show how this analysis can be utilized to select optimal robust policies for an RL-BCI and demonstrate its use on EEG data.

SIGNIFICANCE

We propose here a principled method to determine the optimal policy complexity of an RL problem with a noisy reward, which we argue is particularly useful for RL-based BCI paradigms. This framework may be used to minimize initial training time and allow for a more dynamic and robust shared control between the agent and the operator under different conditions.

A behavioral paradigm for cortical control of a robotic actuator by freely moving rats in a one-dimensional two-target reaching task.. [DOI: 10.1101/2021.01.24.427325] [Citation(s) in RCA: 0] [Impact Index Per Article: 0]

Controlling the trajectory of a neuroprosthetic device to reach multiple targets is a commonly used brain-machine interface (BMI) task in primates and has not been available for rodents yet. Here, we describe a novel experimental paradigm which enables this task for rats in one-dimensional space for reaching two distant targets depending on their limited cognitive and visual capabilities compared to primates. An online transform was used to convert the activity of a pair of primary motor cortex (M1) units into two robotic actions. The rats were shaped to adapt to the transform and direct the robotic actuator toward the selected target by modulating the activity of the M1 neurons. All three rats involved in the study were capable of achieving randomly selected targets with at least 78% accuracy. A total of 9 out of 16 pairs of units examined were eligible for exceeding this success criterion. Two out of three rats were capable of reversal learning, where the mapping between the activity of the unit pairs and the robotic actions were reversed. The present work is the first demonstration of trajectory-based control of a neuroprosthetic device by rodents to reach two distant targets using visual feedback. The paradigm introduced here may be used as a cost-effective platform for elucidating the information processing principles in the neural circuits related to neuroprosthetic control and for studying the performance of novel BMI technologies using freely moving rats.

Neural decoding of continuous upper limb movements: a meta-analysis

OBJECTIVE

EEG-based motion trajectory decoding makes a promising approach for neurotechnology which can be used for neural control of motion reconstruction and neurorehabilitation tools. However, the feasibility and validity of continuous motion decoding by non-invasive brain activity are not clear. The main aim of this study was to perform a meta-analysis across studies that examined the ability of EEG-based continuous motion decoding of upper limb movements.

APPROACH

Pearson's correlation coefficient (CC) was used to evaluate the model performance of the studies and considered as an effect size. To estimate the overall effect size of neural decoding of motion trajectory across studies, characteristics of included studies were addressed and the random effect model was applied to the heterogeneous studies which estimated overall effect size distribution. Furthermore, the significant difference between the two subgroups of imagined and executed movements was analysed.

MAIN RESULTS

The mean of the overall effect size was computed 0.46 across the nonhomogeneous studies. The results showed no significant difference between imagined and executed movements (Chi²=0.28, df = 1, p = 0.60).

SIGNIFICANCE

Meta-analysis results confirm that imagination like execution movements can be used for neural decoding of motion trajectory in neural motor control systems. Also, nonlinear compare with linear model statistically confirmed to be more beneficial for complex movements. Furthermore, a new approach of synergy-based motion decoding can be significantly effective to increase model performance and more research needs to evaluate this method for different levels of complexity of movements.IMPLICATIONS FOR REHABILITATIONNeural decoding methods base on EEG as a non-invasive brain activity, are more user friendly for neurorehabilitation than invasive methods that developing of it makes it more applicable for reconstructing activities of daily living.Neurotechnology for neural control of motion reconstruction, makes the rehabilitation tools to be more synchrony with human intentional movement that can be used to improve the brain neuroplastisity in stroke or other paralysed people.The feasibility and validity of imagined movements equal with executed movements show that amputee people also can benefit EEG-based motion decoding for controling rehabilitation tools just by imagination of their intentional movements.For neurorehabilitation tools, comparing the study outcomes illucidate that the approach of synergy-based motor control in brain activities concluded significantly high performance that highlighted the need it to more investigated in future research.

Deep Learning Based Inter-subject Continuous Decoding of Motor Imagery for Practical Brain-Computer Interfaces

Inter-subject transfer learning is a long-standing problem in brain-computer interfaces (BCIs) and has not yet been fully realized due to high inter-subject variability in the brain signals related to motor imagery (MI). The recent success of deep learning-based algorithms in classifying different brain signals warrants further exploration to determine whether it is feasible for the inter-subject continuous decoding of MI signals to provide contingent neurofeedback which is important for neurorehabilitative BCI designs. In this paper, we have shown how a convolutional neural network (CNN) based deep learning framework can be used for inter-subject continuous decoding of MI related electroencephalographic (EEG) signals using the novel concept of Mega Blocks for adapting the network against inter-subject variabilities. These Mega Blocks have the capacity to repeat a specific architectural block several times such as one or more convolutional layers in a single Mega Block. The parameters of such Mega Blocks can be optimized using Bayesian hyperparameter optimization. The results, obtained on the publicly available BCI competition IV-2b dataset, yields an average inter-subject continuous decoding accuracy of 71.49% (κ = 0.42) and 70.84% (κ = 0.42) for two different training methods such as adaptive moment estimation (Adam) and stochastic gradient descent (SGDM), respectively, in 7 out of 9 subjects. Our results show for the first time that it is feasible to use CNN based architectures for inter-subject continuous decoding with a sufficient level of accuracy for developing calibration-free MI-BCIs for practical purposes.

Reinforcement Learning Based Fast Self-Recalibrating Decoder for Intracortical Brain-Machine Interface

BACKGROUND

For the nonstationarity of neural recordings in intracortical brain-machine interfaces, daily retraining in a supervised manner is always required to maintain the performance of the decoder. This problem can be improved by using a reinforcement learning (RL) based self-recalibrating decoder. However, quickly exploring new knowledge while maintaining a good performance remains a challenge in RL-based decoders.

METHODS

To solve this problem, we proposed an attention-gated RL-based algorithm combining transfer learning, mini-batch, and weight updating schemes to accelerate the weight updating and avoid over-fitting. The proposed algorithm was tested on intracortical neural data recorded from two monkeys to decode their reaching positions and grasping gestures.

RESULTS

The decoding results showed that our proposed algorithm achieved an approximate 20% increase in classification accuracy compared to that obtained by the non-retrained classifier and even achieved better classification accuracy than the daily retraining classifier. Moreover, compared with a conventional RL method, our algorithm improved the accuracy by approximately 10% and the online weight updating speed by approximately 70 times.

CONCLUSIONS

This paper proposed a self-recalibrating decoder which achieved a good and robust decoding performance with fast weight updating and might facilitate its application in wearable device and clinical practice.

A platform for semiautomated voluntary training of common marmosets for behavioral neuroscience

Generally behavioral neuroscience studies of the common marmoset employ adaptations of well-established training methods used with macaque monkeys. However, in many cases these approaches do not readily generalize to marmosets indicating a need for alternatives. Here we present the development of one such alternate: a platform for semiautomated, voluntary in-home cage behavioral training that allows for the study of naturalistic behaviors. We describe the design and production of a modular behavioral training apparatus using CAD software and digital fabrication. We demonstrate that this apparatus permits voluntary behavioral training and data collection throughout the marmoset's waking hours with little experimenter intervention. Furthermore, we demonstrate the use of this apparatus to reconstruct the kinematics of the marmoset's upper limb movement during natural foraging behavior.NEW & NOTEWORTHY The study of marmosets in neuroscience has grown rapidly and presents unique challenges. We address those challenges with an innovative platform for semiautomated, voluntary training that allows marmosets to train throughout their waking hours with minimal experimenter intervention. We describe the use of this platform to capture upper limb kinematics during foraging and to expand the opportunities for behavioral training beyond the limits of traditional training sessions. This flexible platform can easily incorporate other tasks.

Spinal cord neural interfacing in common marmosets (Callithrix jacchus)

Objective:

Spinal cord injury (SCI) remains an ailment with no comprehensive cure, and affected patients suffer from a greatly diminished quality of life. This large population could significantly benefit from prosthetic technologies to replace missing limbs, reanimate nonfunctional limbs, and enable new modes of technologies to restore muscle control and function. While cortically driven brain machine interfaces (BMIs) have achieved great success in interfacing with an external device to restore lost functions, interfacing with the spinal cord can provide an additional site to record motor control signals, which can have its own advantages, albeit challenges from using a smaller non-human primate (NHP) model. The goal of this study is to develop such a spinal cord neural interface to record motor signals from the high cervical levels of the spinal cord in a common marmoset (Callithrix jacchus) model.

Approach and Main Results:

Detailed methods are discussed for this smaller NHP model that includes behavioral training, surgical methods for electrode placement, connector placement and wire handling, electrode specifications and modifications for accessing high cervical level interneurons and motorneurons. The study also discusses the methods and challenges involved in behavioral multi-channel extracellular recording from the marmoset spinal cord, including the major recording failure mechanisms encountered during the study.

Significance:

Marmosets provide a good step between rodent and larger NHP models due to their small size, ease of handling, cognitive abilities, and similarities to other primate motor systems. The study shows the feasibility of recording spinal cord signals and using marmosets as a smaller NHP model in behavioral neuroscience studies. Interfacing with the spinal cord in chronically implanted animals can provide useful information about how motor control signals within the spinal cord are transformed to cause limb movements.

Cortical Topography of Error-Related High-Frequency Potentials During Erroneous Control in a Continuous Control Brain-Computer Interface

Brain–computer interfaces (BCIs) benefit greatly from performance feedback, but current systems lack automatic, task-independent feedback. Cortical responses elicited from user error have the potential to serve as state-based feedback to BCI decoders. To gain a better understanding of local error potentials, we investigate responsive cortical power underlying error-related potentials (ErrPs) from the human cortex during a one-dimensional center-out BCI task, tracking the topography of high-gamma (70–100 Hz) band power (HBP) specific to BCI error. We measured electrocorticography (ECoG) in three human subjects during dynamic, continuous control over BCI cursor velocity. Subjects used motor imagery and rest to move the cursor toward and subsequently dwell within a target region. We then identified and labeled epochs where the BCI decoder incorrectly moved the cursor in the direction opposite of the subject’s expectations (i.e., BCI error). We found increased HBP in various cortical areas 100–500 ms following BCI error with respect to epochs of correct, intended control. Significant responses were noted in primary somatosensory, motor, premotor, and parietal areas and generally regardless of whether the subject was using motor imagery or rest to move the cursor toward the target. Parts of somatosensory, temporal, and parietal areas exclusively had increased HBP when subjects were using motor imagery. In contrast, only part of the parietal cortex near the angular gyrus exclusively had an increase in HBP during rest. This investigation is, to our knowledge, the first to explore cortical fields changes in the context of continuous control in ECoG BCI. We present topographical changes in HBP characteristic specific to the generation of error. By focusing on continuous control, instead of on discrete control for simple selection, we investigate a more naturalistic setting and provide high ecological validity for characterizing error potentials. Such potentials could be considered as design elements for co-adaptive BCIs in the future as task-independent feedback to the decoder, allowing for more robust and individualized BCIs.

A New Frontier: The Convergence of Nanotechnology, Brain Machine Interfaces, and Artificial Intelligence

A confluence of technological capabilities is creating an opportunity for machine learning and artificial intelligence (AI) to enable "smart" nanoengineered brain machine interfaces (BMI). This new generation of technologies will be able to communicate with the brain in ways that support contextual learning and adaptation to changing functional requirements. This applies to both invasive technologies aimed at restoring neurological function, as in the case of neural prosthesis, as well as non-invasive technologies enabled by signals such as electroencephalograph (EEG). Advances in computation, hardware, and algorithms that learn and adapt in a contextually dependent way will be able to leverage the capabilities that nanoengineering offers the design and functionality of BMI. We explore the enabling capabilities that these devices may exhibit, why they matter, and the state of the technologies necessary to build them. We also discuss a number of open technical challenges and problems that will need to be solved in order to achieve this.

Robust Closed-Loop Control of a Cursor in a Person with Tetraplegia using Gaussian Process Regression

Intracortical brain computer interfaces can enable individuals with paralysis to control external devices through voluntarily modulated brain activity. Decoding quality has been previously shown to degrade with signal nonstationarities-specifically, the changes in the statistics of the data between training and testing data sets. This includes changes to the neural tuning profiles and baseline shifts in firing rates of recorded neurons, as well as nonphysiological noise. While progress has been made toward providing long-term user control via decoder recalibration, relatively little work has been dedicated to making the decoding algorithm more resilient to signal nonstationarities. Here, we describe how principled kernel selection with gaussian process regression can be used within a Bayesian filtering framework to mitigate the effects of commonly encountered nonstationarities. Given a supervised training set of (neural features, intention to move in a direction)-pairs, we use gaussian process regression to predict the intention given the neural data. We apply kernel embedding for each neural feature with the standard radial basis function. The multiple kernels are then summed together across each neural dimension, which allows the kernel to effectively ignore large differences that occur only in a single feature. The summed kernel is used for real-time predictions of the posterior mean and variance under a gaussian process framework. The predictions are then filtered using the discriminative Kalman filter to produce an estimate of the neural intention given the history of neural data. We refer to the multiple kernel approach combined with the discriminative Kalman filter as the MK-DKF. We found that the MK-DKF decoder was more resilient to nonstationarities frequently encountered in-real world settings yet provided similar performance to the currently used Kalman decoder. These results demonstrate a method by which neural decoding can be made more resistant to nonstationarities.

Applications of Deep Learning and Reinforcement Learning to Biological Data

Rapid advances in hardware-based technologies during the past decades have opened up new possibilities for life scientists to gather multimodal data in various application domains, such as omics, bioimaging, medical imaging, and (brain/body)-machine interfaces. These have generated novel opportunities for development of dedicated data-intensive machine learning techniques. In particular, recent research in deep learning (DL), reinforcement learning (RL), and their combination (deep RL) promise to revolutionize the future of artificial intelligence. The growth in computational power accompanied by faster and increased data storage, and declining computing costs have already allowed scientists in various fields to apply these techniques on data sets that were previously intractable owing to their size and complexity. This paper provides a comprehensive survey on the application of DL, RL, and deep RL techniques in mining biological data. In addition, we compare the performances of DL techniques when applied to different data sets across various application domains. Finally, we outline open issues in this challenging research area and discuss future development perspectives.

Progress in Neuroengineering for brain repair: New challenges and open issues

Background

In recent years, biomedical devices have proven to be able to target also different neurological disorders. Given the rapid ageing of the population and the increase of invalidating diseases affecting the central nervous system, there is a growing demand for biomedical devices of immediate clinical use. However, to reach useful therapeutic results, these tools need a multidisciplinary approach and a continuous dialogue between neuroscience and engineering, a field that is named neuroengineering. This is because it is fundamental to understand how to read and perturb the neural code in order to produce a significant clinical outcome.

Results

In this review, we first highlight the importance of developing novel neurotechnological devices for brain repair and the major challenges expected in the next years. We describe the different types of brain repair strategies being developed in basic and clinical research and provide a brief overview of recent advances in artificial intelligence that have the potential to improve the devices themselves. We conclude by providing our perspective on their implementation to humans and the ethical issues that can arise.

Conclusions

Neuroengineering approaches promise to be at the core of future developments for clinical applications in brain repair, where the boundary between biology and artificial intelligence will become increasingly less pronounced.

Augmenting intracortical brain-machine interface with neurally driven error detectors

OBJECTIVE

Making mistakes is inevitable, but identifying them allows us to correct or adapt our behavior to improve future performance. Current brain-machine interfaces (BMIs) make errors that need to be explicitly corrected by the user, thereby consuming time and thus hindering performance. We hypothesized that neural correlates of the user perceiving the mistake could be used by the BMI to automatically correct errors. However, it was unknown whether intracortical outcome error signals were present in the premotor and primary motor cortices, brain regions successfully used for intracortical BMIs.

APPROACH

We report here for the first time a putative outcome error signal in spiking activity within these cortices when rhesus macaques performed an intracortical BMI computer cursor task.

MAIN RESULTS

We decoded BMI trial outcomes shortly after and even before a trial ended with 96% and 84% accuracy, respectively. This led us to develop and implement in real-time a first-of-its-kind intracortical BMI error 'detect-and-act' system that attempts to automatically 'undo' or 'prevent' mistakes. The detect-and-act system works independently and in parallel to a kinematic BMI decoder. In a challenging task that resulted in substantial errors, this approach improved the performance of a BMI employing two variants of the ubiquitous Kalman velocity filter, including a state-of-the-art decoder (ReFIT-KF).

SIGNIFICANCE

Detecting errors in real-time from the same brain regions that are commonly used to control BMIs should improve the clinical viability of BMIs aimed at restoring motor function to people with paralysis.

Feasibility of Automatic Error Detect-and-Undo System in Human Intracortical Brain-Computer Interfaces

OBJECTIVE

Brain-computer interfaces (BCIs) aim to help people with impaired movement ability by directly translating their movement intentions into command signals for assistive technologies. Despite large performance improvements over the last two decades, BCI systems still make errors that need to be corrected manually by the user. This decreases system performance and is also frustrating for the user. The deleterious effects of errors could be mitigated if the system automatically detected when the user perceives that an error was made and automatically intervened with a corrective action; thus, sparing users from having to make the correction themselves. Our previous preclinical work with monkeys demonstrated that task-outcome correlates exist in motor cortical spiking activity and can be utilized to improve BCI performance. Here, we asked if these signals also exist in the human hand area of motor cortex, and whether they can be decoded with high accuracy.

METHODS

We analyzed posthoc the intracortical neural activity of two BrainGate2 clinical trial participants who were neurally controlling a computer cursor to perform a grid target selection task and a keyboard-typing task.

RESULTS

Our key findings are that: 1) there exists a putative outcome error signal reflected in both the action potentials and local field potentials of the human hand area of motor cortex, and 2) target selection outcomes can be classified with high accuracy (70-85%) of errors successfully detected with minimal (0-3%) misclassifications of success trials, based on neural activity alone.

SIGNIFICANCE

These offline results suggest that it will be possible to improve the performance of clinical intracortical BCIs by incorporating a real-time error detect-and-undo system alongside the decoding of movement intention.

Leveraging neural dynamics to extend functional lifetime of brain-machine interfaces

Intracortical brain-machine interfaces (BMIs) aim to restore lost motor function to people with neurological deficits by decoding neural activity into control signals for guiding prostheses. An important challenge facing BMIs is that, over time, the number of neural signals recorded from implanted multielectrode arrays will decline and result in a concomitant decrease of BMI performance. We sought to extend BMI lifetime by developing an algorithmic technique, implemented entirely in software, to improve performance over state-of-the-art algorithms as the number of recorded neural signals decline. Our approach augments the decoder by incorporating neural population dynamics remembered from an earlier point in the array lifetime. We demonstrate, in closed-loop experiments with two rhesus macaques, that after the loss of approximately 60% of recording electrodes, our approach outperforms state-of-the-art decoders by a factor of 3.2× and 1.7× (corresponding to a 46% and 22% recovery of maximal performance). Further, our results suggest that neural population dynamics in motor cortex are invariant to the number of recorded neurons. By extending functional BMI lifetime, this approach increases the clinical viability of BMIs.

Brain-Machine Interfaces: From Basic Science to Neuroprostheses and Neurorehabilitation

Brain-machine interfaces (BMIs) combine methods, approaches, and concepts derived from neurophysiology, computer science, and engineering in an effort to establish real-time bidirectional links between living brains and artificial actuators. Although theoretical propositions and some proof of concept experiments on directly linking the brains with machines date back to the early 1960s, BMI research only took off in earnest at the end of the 1990s, when this approach became intimately linked to new neurophysiological methods for sampling large-scale brain activity. The classic goals of BMIs are 1) to unveil and utilize principles of operation and plastic properties of the distributed and dynamic circuits of the brain and 2) to create new therapies to restore mobility and sensations to severely disabled patients. Over the past decade, a wide range of BMI applications have emerged, which considerably expanded these original goals. BMI studies have shown neural control over the movements of robotic and virtual actuators that enact both upper and lower limb functions. Furthermore, BMIs have also incorporated ways to deliver sensory feedback, generated from external actuators, back to the brain. BMI research has been at the forefront of many neurophysiological discoveries, including the demonstration that, through continuous use, artificial tools can be assimilated by the primate brain's body schema. Work on BMIs has also led to the introduction of novel neurorehabilitation strategies. As a result of these efforts, long-term continuous BMI use has been recently implicated with the induction of partial neurological recovery in spinal cord injury patients.

Common marmoset (Callithrix jacchus) as a primate model for behavioral neuroscience studies

BACKGROUND

The common marmoset (Callithrix jacchus) has been proposed as a suitable bridge between rodents and larger primates. They have been used in several types of research including auditory, vocal, visual, pharmacological and genetics studies. However, marmosets have not been used as much for behavioral studies.

NEW METHOD

Here we present data from training 12 adult marmosets for behavioral neuroscience studies. We discuss the husbandry, food preferences, handling, acclimation to laboratory environments and neurosurgical techniques. In this paper, we also present a custom built "scoop" and a monkey chair suitable for training of these animals.

RESULTS

The animals were trained for three tasks: 4 target center-out reaching task, reaching tasks that involved controlling robot actions, and touch screen task. All animals learned the center-out reaching task within 1-2 weeks whereas learning reaching tasks controlling robot actions task took several months of behavioral training where the monkeys learned to associate robot actions with food rewards.

COMPARISON TO EXISTING METHOD

We propose the marmoset as a novel model for behavioral neuroscience research as an alternate for larger primate models. This is due to the ease of handling, quick reproduction, available neuroanatomy, sensorimotor system similar to larger primates and humans, and a lissencephalic brain that can enable implantation of microelectrode arrays relatively easier at various cortical locations compared to larger primates.

CONCLUSION

All animals were able to learn behavioral tasks well and we present the marmosets as an alternate model for simple behavioral neuroscience tasks.

Feedback for reinforcement learning based brain-machine interfaces using confidence metrics

OBJECTIVE

For brain-machine interfaces (BMI) to be used in activities of daily living by paralyzed individuals, the BMI should be as autonomous as possible. One of the challenges is how the feedback is extracted and utilized in the BMI. Our long-term goal is to create autonomous BMIs that can utilize an evaluative feedback from the brain to update the decoding algorithm and use it intelligently in order to adapt the decoder. In this study, we show how to extract the necessary evaluative feedback from a biologically realistic (synthetic) source, use both the quantity and the quality of the feedback, and how that feedback information can be incorporated into a reinforcement learning (RL) controller architecture to maximize its performance.

APPROACH

Motivated by the perception-action-reward cycle (PARC) in the brain which links reward for cognitive decision making and goal-directed behavior, we used a reward-based RL architecture named Actor-Critic RL as the model. Instead of using an error signal towards building an autonomous BMI, we envision to use a reward signal from the nucleus accumbens (NAcc) which plays a key role in the linking of reward to motor behaviors. To deal with the complexity and non-stationarity of biological reward signals, we used a confidence metric which was used to indicate the degree of feedback accuracy. This confidence was added to the Actor's weight update equation in the RL controller architecture. If the confidence was high (>0.2), the BMI decoder used this feedback to update its parameters. However, when the confidence was low, the BMI decoder ignored the feedback and did not update its parameters. The range between high confidence and low confidence was termed as the 'ambiguous' region. When the feedback was within this region, the BMI decoder updated its weight at a lower rate than when fully confident, which was decided by the confidence. We used two biologically realistic models to generate synthetic data for MI (Izhikevich model) and NAcc (Humphries model) to validate proposed controller architecture.

MAIN RESULTS

In this work, we show how the overall performance of the BMI was improved by using a threshold close to the decision boundary to reject erroneous feedback. Additionally, we show the stability of the system improved when the feedback was used with a threshold.

SIGNIFICANCE

The result of this study is a step towards making BMIs autonomous. While our method is not fully autonomous, the results demonstrate that extensive training times necessary at the beginning of each BMI session can be significantly decreased. In our approach, decoder training time was only limited to 10 trials in the first BMI session. Subsequent sessions used previous session weights to initialize the decoder. We also present a method where the use of a threshold can be applied to any decoder with a feedback signal that is less than perfect so that erroneous feedback can be avoided and the stability of the system can be increased.

A Sliced Inverse Regression (SIR) Decoding the Forelimb Movement from Neuronal Spikes in the Rat Motor Cortex

Several neural decoding algorithms have successfully converted brain signals into commands to control a computer cursor and prosthetic devices. A majority of decoding methods, such as population vector algorithms (PVA), optimal linear estimators (OLE), and neural networks (NN), are effective in predicting movement kinematics, including movement direction, speed and trajectory but usually require a large number of neurons to achieve desirable performance. This study proposed a novel decoding algorithm even with signals obtained from a smaller numbers of neurons. We adopted sliced inverse regression (SIR) to predict forelimb movement from single-unit activities recorded in the rat primary motor (M1) cortex in a water-reward lever-pressing task. SIR performed weighted principal component analysis (PCA) to achieve effective dimension reduction for nonlinear regression. To demonstrate the decoding performance, SIR was compared to PVA, OLE, and NN. Furthermore, PCA and sequential feature selection (SFS) which are popular feature selection techniques were implemented for comparison of feature selection effectiveness. Among SIR, PVA, OLE, PCA, SFS, and NN decoding methods, the trajectories predicted by SIR (with a root mean square error, RMSE, of 8.47 ± 1.32 mm) was closer to the actual trajectories compared with those predicted by PVA (30.41 ± 11.73 mm), OLE (20.17 ± 6.43 mm), PCA (19.13 ± 0.75 mm), SFS (22.75 ± 2.01 mm), and NN (16.75 ± 2.02 mm). The superiority of SIR was most obvious when the sample size of neurons was small. We concluded that SIR sorted the input data to obtain the effective transform matrices for movement prediction, making it a robust decoding method for conditions with sparse neuronal information.

Making brain-machine interfaces robust to future neural variability

A major hurdle to clinical translation of brain–machine interfaces (BMIs) is that current decoders, which are trained from a small quantity of recent data, become ineffective when neural recording conditions subsequently change. We tested whether a decoder could be made more robust to future neural variability by training it to handle a variety of recording conditions sampled from months of previously collected data as well as synthetic training data perturbations. We developed a new multiplicative recurrent neural network BMI decoder that successfully learned a large variety of neural-to-kinematic mappings and became more robust with larger training data sets. Here we demonstrate that when tested with a non-human primate preclinical BMI model, this decoder is robust under conditions that disabled a state-of-the-art Kalman filter-based decoder. These results validate a new BMI strategy in which accumulated data history are effectively harnessed, and may facilitate reliable BMI use by reducing decoder retraining downtime.

Brain-machine interfaces (BMI) depend on algorithms to decode neural signals, but these decoders cope poorly with signal variability. Here, authors report a BMI decoder which circumvents these problems by using a large and perturbed training dataset to improve performance with variable neural signals.

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.

Neuroprosthetic Decoder Training as Imitation Learning

Neuroprosthetic brain-computer interfaces function via an algorithm which decodes neural activity of the user into movements of an end effector, such as a cursor or robotic arm. In practice, the decoder is often learned by updating its parameters while the user performs a task. When the user's intention is not directly observable, recent methods have demonstrated value in training the decoder against a surrogate for the user's intended movement. Here we show that training a decoder in this way is a novel variant of an imitation learning problem, where an oracle or expert is employed for supervised training in lieu of direct observations, which are not available. Specifically, we describe how a generic imitation learning meta-algorithm, dataset aggregation (DAgger), can be adapted to train a generic brain-computer interface. By deriving existing learning algorithms for brain-computer interfaces in this framework, we provide a novel analysis of regret (an important metric of learning efficacy) for brain-computer interfaces. This analysis allows us to characterize the space of algorithmic variants and bounds on their regret rates. Existing approaches for decoder learning have been performed in the cursor control setting, but the available design principles for these decoders are such that it has been impossible to scale them to naturalistic settings. Leveraging our findings, we then offer an algorithm that combines imitation learning with optimal control, which should allow for training of arbitrary effectors for which optimal control can generate goal-oriented control. We demonstrate this novel and general BCI algorithm with simulated neuroprosthetic control of a 26 degree-of-freedom model of an arm, a sophisticated and realistic end effector.

Overcoming Long-Term Variability in Local Field Potentials Using an Adaptive Decoder

Long-term variability remains one of the major hurdles in using intracortical recordings like local field potentials for brain computer interfaces (BCI). Practical neural decoders need to overcome time instability of neural signals to estimate subject behavior accurately and faithfully over the long term. This paper presents a novel decoder that 1) characterizes each behavioral task (i.e., different movement directions under different force conditions) with multiple neural patterns and 2) adapts to the long-term variations in neural features by identifying the stable neural patterns. This adaptation can be performed in both an unsupervised and a semisupervised learning framework requiring minimal feedback from the user. To achieve generalization over time, the proposed decoder uses redundant sparse regression models that adapt to day-to-day variations in neural patterns. While this update requires no explicit feedback from the BCI user, any feedback (explicit or derived) to the BCI improves its performance. With this adaptive decoder, we investigated the effects of long-term neural modulation especially when subjects encountered new external forces against movement. The proposed decoder predicted eight hand-movement directions with an accuracy of 95% over two weeks (when there was no external forces); and 85% in later acquisition sessions spanning up to 42 days (when the monkeys countered external field forces). Since the decoder can operate with or without manual intervention, it could alleviate user frustration associated with BCI.

Brain-state classification and a dual-state decoder dramatically improve the control of cursor movement through a brain-machine interface

OBJECTIVE

It is quite remarkable that brain machine interfaces (BMIs) can be used to control complex movements with fewer than 100 neurons. Success may be due in part to the limited range of dynamical conditions under which most BMIs are tested. Achieving high-quality control that spans these conditions with a single linear mapping will be more challenging. Even for simple reaching movements, existing BMIs must reduce the stochastic noise of neurons by averaging the control signals over time, instead of over the many neurons that normally control movement. This forces a compromise between a decoder with dynamics allowing rapid movement and one that allows postures to be maintained with little jitter. Our current work presents a method for addressing this compromise, which may also generalize to more highly varied dynamical situations, including movements with more greatly varying speed.

APPROACH

We have developed a system that uses two independent Wiener filters as individual components in a single decoder, one optimized for movement, and the other for postural control. We computed an LDA classifier using the same neural inputs. The decoder combined the outputs of the two filters in proportion to the likelihood assigned by the classifier to each state.

MAIN RESULTS

We have performed online experiments with two monkeys using this neural-classifier, dual-state decoder, comparing it to a standard, single-state decoder as well as to a dual-state decoder that switched states automatically based on the cursor's proximity to a target. The performance of both monkeys using the classifier decoder was markedly better than that of the single-state decoder and comparable to the proximity decoder.

SIGNIFICANCE

We have demonstrated a novel strategy for dealing with the need to make rapid movements while also maintaining precise cursor control when approaching and stabilizing within targets. Further gains can undoubtedly be realized by optimizing the performance of the individual movement and posture decoders.

Assessing Granger Causality in Electrophysiological Data: Removing the Adverse Effects of Common Signals via Bipolar Derivations

Multielectrode voltage data are usually recorded against a common reference. Such data are frequently used without further treatment to assess patterns of functional connectivity between neuronal populations and between brain areas. It is important to note from the outset that such an approach is valid only when the reference electrode is nearly electrically silent. In practice, however, the reference electrode is generally not electrically silent, thereby adding a common signal to the recorded data. Volume conduction further complicates the problem. In this study we demonstrate the adverse effects of common signals on the estimation of Granger causality, which is a statistical measure used to infer synaptic transmission and information flow in neural circuits from multielectrode data. We further test the hypothesis that the problem can be overcome by utilizing bipolar derivations where the difference between two nearby electrodes is taken and treated as a representation of local neural activity. Simulated data generated by a neuronal network model where the connectivity pattern is known were considered first. This was followed by analyzing data from three experimental preparations where a priori predictions regarding the patterns of causal interactions can be made: (1) laminar recordings from the hippocampus of an anesthetized rat during theta rhythm, (2) laminar recordings from V4 of an awake-behaving macaque monkey during alpha rhythm, and (3) ECoG recordings from electrode arrays implanted in the middle temporal lobe and prefrontal cortex of an epilepsy patient during fixation. For both simulation and experimental analysis the results show that bipolar derivations yield the expected connectivity patterns whereas the untreated data (referred to as unipolar signals) do not. In addition, current source density signals, where applicable, yield results that are close to the expected connectivity patterns, whereas the commonly practiced average re-reference method leads to erroneous results.

Toward Building Hybrid Biological/in silico Neural Networks for Motor Neuroprosthetic Control

In this article, we introduce the Bioinspired Neuroprosthetic Design Environment (BNDE) as a practical platform for the development of novel brain–machine interface (BMI) controllers, which are based on spiking model neurons. We built the BNDE around a hard real-time system so that it is capable of creating simulated synapses from extracellularly recorded neurons to model neurons. In order to evaluate the practicality of the BNDE for neuroprosthetic control experiments, a novel, adaptive BMI controller was developed and tested using real-time closed-loop simulations. The present controller consists of two in silico medium spiny neurons, which receive simulated synaptic inputs from recorded motor cortical neurons. In the closed-loop simulations, the recordings from the cortical neurons were imitated using an external, hardware-based neural signal synthesizer. By implementing a reward-modulated spike timing-dependent plasticity rule, the controller achieved perfect target reach accuracy for a two-target reaching task in one-dimensional space. The BNDE combines the flexibility of software-based spiking neural network (SNN) simulations with powerful online data visualization tools and is a low-cost, PC-based, and all-in-one solution for developing neurally inspired BMI controllers. We believe that the BNDE is the first implementation, which is capable of creating hybrid biological/in silico neural networks for motor neuroprosthetic control and utilizes multiple CPU cores for computationally intensive real-time SNN simulations.

Toward an autonomous brain machine interface: integrating sensorimotor reward modulation and reinforcement learning

For decades, neurophysiologists have worked on elucidating the function of the cortical sensorimotor control system from the standpoint of kinematics or dynamics. Recently, computational neuroscientists have developed models that can emulate changes seen in the primary motor cortex during learning. However, these simulations rely on the existence of a reward-like signal in the primary sensorimotor cortex. Reward modulation of the primary sensorimotor cortex has yet to be characterized at the level of neural units. Here we demonstrate that single units/multiunits and local field potentials in the primary motor (M1) cortex of nonhuman primates (Macaca radiata) are modulated by reward expectation during reaching movements and that this modulation is present even while subjects passively view cursor motions that are predictive of either reward or nonreward. After establishing this reward modulation, we set out to determine whether we could correctly classify rewarding versus nonrewarding trials, on a moment-to-moment basis. This reward information could then be used in collaboration with reinforcement learning principles toward an autonomous brain-machine interface. The autonomous brain-machine interface would use M1 for both decoding movement intention and extraction of reward expectation information as evaluative feedback, which would then update the decoding algorithm as necessary. In the work presented here, we show that this, in theory, is possible.

Kernel temporal differences for neural decoding

We study the feasibility and capability of the kernel temporal difference (KTD)(λ) algorithm for neural decoding. KTD(λ) is an online, kernel-based learning algorithm, which has been introduced to estimate value functions in reinforcement learning. This algorithm combines kernel-based representations with the temporal difference approach to learning. One of our key observations is that by using strictly positive definite kernels, algorithm's convergence can be guaranteed for policy evaluation. The algorithm's nonlinear functional approximation capabilities are shown in both simulations of policy evaluation and neural decoding problems (policy improvement). KTD can handle high-dimensional neural states containing spatial-temporal information at a reasonable computational complexity allowing real-time applications. When the algorithm seeks a proper mapping between a monkey's neural states and desired positions of a computer cursor or a robot arm, in both open-loop and closed-loop experiments, it can effectively learn the neural state to action mapping. Finally, a visualization of the coadaptation process between the decoder and the subject shows the algorithm's capabilities in reinforcement learning brain machine interfaces.

Decoding methods for neural prostheses: where have we reached?

This article reviews advances in decoding methods for brain-machine interfaces (BMIs). Recent work has focused on practical considerations for future clinical deployment of prosthetics. This review is organized by open questions in the field such as what variables to decode, how to design neural tuning models, which neurons to select, how to design models of desired actions, how to learn decoder parameters during prosthetic operation, and how to adapt to changes in neural signals and neural tuning. The concluding discussion highlights the need to design and test decoders within the context of their expected use and the need to answer the question of how much control accuracy is good enough for a prosthetic.

Patterns of afferent input to the caudal and rostral areas of the dorsal premotor cortex (6DC and 6DR) in the marmoset monkey

Corticocortical projections to the caudal and rostral areas of dorsal premotor cortex (6DC and 6DR, also known as F2 and F7) were studied in the marmoset monkey. Both areas received their main thalamic inputs from the ventral anterior and ventral lateral complexes, and received dense projections from the medial premotor cortex. However, there were marked differences in their connections with other cortical areas. While 6DR received consistent inputs from prefrontal cortex, area 6DC received few such connections. Conversely, 6DC, but not 6DR, received major projections from the primary motor and somatosensory areas. Projections from the anterior cingulate cortex preferentially targeted 6DC, while the posterior cingulate and adjacent medial wall areas preferentially targeted 6DR. Projections from the medial parietal area PE to 6DC were particularly dense, while intraparietal areas (especially the putative homolog of LIP) were more strongly labeled after 6DR injections. Finally, 6DC and 6DR were distinct in terms of inputs from the ventral parietal cortex: projections to 6DR originated preferentially from caudal areas (PG and OPt), while 6DC received input primarily from rostral areas (PF and PFG). Differences in connections suggest that area 6DR includes rostral and caudal subdivisions, with the former also involved in oculomotor control. These results suggest that area 6DC is more directly involved in the preparation and execution of motor acts, while area 6DR integrates sensory and internally driven inputs for the planning of goal-directed actions. They also provide strong evidence of a homologous organization of the dorsal premotor cortex in New and Old World monkeys.

A confidence metric for using neurobiological feedback in actor-critic reinforcement learning based brain-machine interfaces

Brain-Machine Interfaces (BMIs) can be used to restore function in people living with paralysis. Current BMIs require extensive calibration that increase the set-up times and external inputs for decoder training that may be difficult to produce in paralyzed individuals. Both these factors have presented challenges in transitioning the technology from research environments to activities of daily living (ADL). For BMIs to be seamlessly used in ADL, these issues should be handled with minimal external input thus reducing the need for a technician/caregiver to calibrate the system. Reinforcement Learning (RL) based BMIs are a good tool to be used when there is no external training signal and can provide an adaptive modality to train BMI decoders. However, RL based BMIs are sensitive to the feedback provided to adapt the BMI. In actor-critic BMIs, this feedback is provided by the critic and the overall system performance is limited by the critic accuracy. In this work, we developed an adaptive BMI that could handle inaccuracies in the critic feedback in an effort to produce more accurate RL based BMIs. We developed a confidence measure, which indicated how appropriate the feedback is for updating the decoding parameters of the actor. The results show that with the new update formulation, the critic accuracy is no longer a limiting factor for the overall performance. We tested and validated the system onthree different data sets: synthetic data generated by an Izhikevich neural spiking model, synthetic data with a Gaussian noise distribution, and data collected from a non-human primate engaged in a reaching task. All results indicated that the system with the critic confidence built in always outperformed the system without the critic confidence. Results of this study suggest the potential application of the technique in developing an autonomous BMI that does not need an external signal for training or extensive calibration.