The science and engineering behind sensitized brain-controlled bionic hands

Reducing power requirements for high-accuracy decoding in iBCIs

Objective.Current intracortical brain-computer interfaces (iBCIs) rely predominantly on threshold crossings ('spikes') for decoding neural activity into a control signal for an external device. Spiking data can yield high accuracy online control during complex behaviors; however, its dependence on high-sampling-rate data collection can pose challenges. An alternative signal for iBCI decoding is the local field potential (LFP), a continuous-valued signal that can be acquired simultaneously with spiking activity. However, LFPs are seldom used alone for online iBCI control as their decoding performance has yet to achieve parity with spikes.Approach.Here, we present a strategy to improve the performance of LFP-based decoders by first training a neural dynamics model to use LFPs to reconstruct the firing rates underlying spiking data, and then decoding from the estimated rates. We test these models on previously-collected macaque data during center-out and random-target reaching tasks as well as data collected from a human iBCI participant during attempted speech.Main results.In all cases, training models from LFPs enables firing rate reconstruction with accuracy comparable to spiking-based dynamics models. In addition, LFP-based dynamics models enable decoding performance exceeding that of LFPs alone and approaching that of spiking-based models. In all applications except speech, LFP-based dynamics models also facilitate decoding accuracy exceeding that of direct decoding from spikes.Significance.Because LFP-based dynamics models operate on lower bandwidth and with lower sampling rate than spiking models, our findings indicate that iBCI devices can be designed to operate with lower power requirements than devices dependent on recorded spiking activity, without sacrificing high-accuracy decoding.

Few-shot Algorithms for Consistent Neural Decoding (FALCON) Benchmark

Intracortical brain-computer interfaces (iBCIs) can restore movement and communication abilities to individuals with paralysis by decoding their intended behavior from neural activity recorded with an implanted device. While this activity yields high-performance decoding over short timescales, neural data are often nonstationary, which can lead to decoder failure if not accounted for. To maintain performance, users must frequently recalibrate decoders, which requires the arduous collection of new neural and behavioral data. Aiming to reduce this burden, several approaches have been developed that either limit recalibration data requirements (few-shot approaches) or eliminate explicit recalibration entirely (zero-shot approaches). However, progress is limited by a lack of standardized datasets and comparison metrics, causing methods to be compared in an ad hoc manner. Here we introduce the FALCON benchmark suite (Few-shot Algorithms for COnsistent Neural decoding) to standardize evaluation of iBCI robustness. FALCON curates five datasets of neural and behavioral data that span movement and communication tasks to focus on behaviors of interest to modern-day iBCIs. Each dataset includes calibration data, optional few-shot recalibration data, and private evaluation data. We implement a flexible evaluation platform which only requires user-submitted code to return behavioral predictions on unseen data. We also seed the benchmark by applying baseline methods spanning several classes of possible approaches. FALCON aims to provide rigorous selection criteria for robust iBCI decoders, easing their translation to real-world devices.

Sensory integration for neuroprostheses: from functional benefits to neural correlates

In the field of sensory neuroprostheses, one ultimate goal is for individuals to perceive artificial somatosensory information and use the prosthesis with high complexity that resembles an intact system. To this end, research has shown that stimulation-elicited somatosensory information improves prosthesis perception and task performance. While studies strive to achieve sensory integration, a crucial phenomenon that entails naturalistic interaction with the environment, this topic has not been commensurately reviewed. Therefore, here we present a perspective for understanding sensory integration in neuroprostheses. First, we review the engineering aspects and functional outcomes in sensory neuroprosthesis studies. In this context, we summarize studies that have suggested sensory integration. We focus on how they have used stimulation-elicited percepts to maximize and improve the reliability of somatosensory information. Next, we review studies that have suggested multisensory integration. These works have demonstrated that congruent and simultaneous multisensory inputs provided cognitive benefits such that an individual experiences a greater sense of authority over prosthesis movements (i.e., agency) and perceives the prosthesis as part of their own (i.e., ownership). Thereafter, we present the theoretical and neuroscience framework of sensory integration. We investigate how behavioral models and neural recordings have been applied in the context of sensory integration. Sensory integration models developed from intact-limb individuals have led the way to sensory neuroprosthesis studies to demonstrate multisensory integration. Neural recordings have been used to show how multisensory inputs are processed across cortical areas. Lastly, we discuss some ongoing research and challenges in achieving and understanding sensory integration in sensory neuroprostheses. Resolving these challenges would help to develop future strategies to improve the sensory feedback of a neuroprosthetic system.

Nonlinear model predictive control of a conductance-based neuron model via data-driven forecasting

Objective. Precise control of neural systems is essential to experimental investigations of how the brain controls behavior and holds the potential for therapeutic manipulations to correct aberrant network states. Model predictive control, which employs a dynamical model of the system to find optimal control inputs, has promise for dealing with the nonlinear dynamics, high levels of exogenous noise, and limited information about unmeasured states and parameters that are common in a wide range of neural systems. However, the challenge still remains of selecting the right model, constraining its parameters, and synchronizing to the neural system.Approach. As a proof of principle, we used recent advances in data-driven forecasting to construct a nonlinear machine-learning model of a Hodgkin-Huxley type neuron when only the membrane voltage is observable and there are an unknown number of intrinsic currents.Main Results. We show that this approach is able to learn the dynamics of different neuron types and can be used with model predictive control (MPC) to force the neuron to engage in arbitrary, researcher-defined spiking behaviors.Significance.To the best of our knowledge, this is the first application of nonlinear MPC of a conductance-based model where there is only realistically limited information about unobservable states and parameters.

Somatotopically Evoked Tactile Sensation via Transcutaneous Electrical Nerve Stimulation Improves Prosthetic Sensorimotor Performance

Sensory feedback provides critical interactive information for the effective use of hand prostheses. Non-invasive neural interfaces allow convenient access to the sensory system, but they communicate a limited amount of sensory information. This study examined a novel approach that leverages a direct and natural sensory afferent pathway, and enables an evoked tactile sensation (ETS) of multiple digits in the projected finger map (PFM) of participants with forearm amputation non-invasively. A bidirectional prosthetic interface was constructed by integrating the non-invasive ETS-based feedback system into a commercial prosthetic hand. The pressure information of five fingers was encoded linearly by the pulse width modulation range of the buzz sensation. We showed that simultaneous perception of multiple digits allowed participants with forearm amputation to identify object length and compliance by using information about contact patterns and force intensity. The ETS enhanced the grasp-and-transport performance of participants with and without prior experience of prosthetic use. The functional test of transport-and-identification further revealed improved execution in classifying object size and compliance using ETS-based feedback. Results demonstrated that the ETS is capable of communicating somatotopically compatible information to participants efficiently, and improves sensory discrimination and closed-loop prosthetic control. This non-invasive sensory interface may establish a viable way to restore sensory ability for prosthetic users who experience the phenomenon of PFM.

Evolution, biomechanics, and neurobiology converge to explain selective finger motor control

Humans use their fingers to perform a variety of tasks, from simple grasping to manipulating objects, to typing and playing musical instruments, a variety wider than any other species. The more sophisticated the task, the more it involves individuated finger movements, those in which one or more selected fingers perform an intended action while the motion of other digits is constrained. Here we review the neurobiology of such individuated finger movements. We consider their evolutionary origins, the extent to which finger movements are in fact individuated, and the evolved features of neuromuscular control that both enable and limit individuation. We go on to discuss other features of motor control that combine with individuation to create dexterity, the impairment of individuation by disease, and the broad extent of capabilities that individuation confers on humans. We comment on the challenges facing the development of a truly dexterous bionic hand. We conclude by identifying topics for future investigation that will advance our understanding of how neural networks interact across multiple regions of the central nervous system to create individuated movements for the skills humans use to express their cognitive activity.

Identifying Distinct Neural Features between the Initial and Corrective Phases of Precise Reaching Using AutoLFADS

Many initial movements require subsequent corrective movements, but how the motor cortex transitions to make corrections and how similar the encoding is to initial movements is unclear. In our study, we explored how the brain's motor cortex signals both initial and corrective movements during a precision reaching task. We recorded a large population of neurons from two male rhesus macaques across multiple sessions to examine the neural firing rates during not only initial movements but also subsequent corrective movements. AutoLFADS, an autoencoder-based deep-learning model, was applied to provide a clearer picture of neurons' activity on individual corrective movements across sessions. Decoding of reach velocity generalized poorly from initial to corrective submovements. Unlike initial movements, it was challenging to predict the velocity of corrective movements using traditional linear methods in a single, global neural space. We identified several locations in the neural space where corrective submovements originated after the initial reaches, signifying firing rates different than the baseline before initial movements. To improve corrective movement decoding, we demonstrate that a state-dependent decoder incorporating the population firing rates at the initiation of correction improved performance, highlighting the diverse neural features of corrective movements. In summary, we show neural differences between initial and corrective submovements and how the neural activity encodes specific combinations of velocity and position. These findings are inconsistent with assumptions that neural correlations with kinematic features are global and independent, emphasizing that traditional methods often fall short in describing these diverse neural processes for online corrective movements.

Visual context affects the perceived timing of tactile sensations elicited through intra-cortical microstimulation

Intra-cortical microstimulation (ICMS) is a technique to provide tactile sensations for a somatosensory brain-machine interface (BMI). A viable BMI must function within the rich, multisensory environment of the real world, but how ICMS is integrated with other sensory modalities is poorly understood. To investigate how ICMS percepts are integrated with visual information, ICMS and visual stimuli were delivered at varying times relative to one another. Both visual context and ICMS current amplitude were found to bias the qualitative experience of ICMS. In two tetraplegic participants, ICMS and visual stimuli were more likely to be experienced as occurring simultaneously when visual stimuli were more realistic, demonstrating an effect of visual context on the temporal binding window. The peak of the temporal binding window varied but was consistently offset from zero, suggesting that multisensory integration with ICMS can suffer from temporal misalignment. Recordings from primary somatosensory cortex (S1) during catch trials where visual stimuli were delivered without ICMS demonstrated that S1 represents visual information related to ICMS across visual contexts.

Nonlinear manifolds underlie neural population activity during behaviour

There is rich variety in the activity of single neurons recorded during behaviour. Yet, these diverse single neuron responses can be well described by relatively few patterns of neural co-modulation. The study of such low-dimensional structure of neural population activity has provided important insights into how the brain generates behaviour. Virtually all of these studies have used linear dimensionality reduction techniques to estimate these population-wide co-modulation patterns, constraining them to a flat "neural manifold". Here, we hypothesised that since neurons have nonlinear responses and make thousands of distributed and recurrent connections that likely amplify such nonlinearities, neural manifolds should be intrinsically nonlinear. Combining neural population recordings from monkey, mouse, and human motor cortex, and mouse striatum, we show that: 1) neural manifolds are intrinsically nonlinear; 2) their nonlinearity becomes more evident during complex tasks that require more varied activity patterns; and 3) manifold nonlinearity varies across architecturally distinct brain regions. Simulations using recurrent neural network models confirmed the proposed relationship between circuit connectivity and manifold nonlinearity, including the differences across architecturally distinct regions. Thus, neural manifolds underlying the generation of behaviour are inherently nonlinear, and properly accounting for such nonlinearities will be critical as neuroscientists move towards studying numerous brain regions involved in increasingly complex and naturalistic behaviours.

BRAND: a platform for closed-loop experiments with deep network models

Objective.Artificial neural networks (ANNs) are state-of-the-art tools for modeling and decoding neural activity, but deploying them in closed-loop experiments with tight timing constraints is challenging due to their limited support in existing real-time frameworks. Researchers need a platform that fully supports high-level languages for running ANNs (e.g. Python and Julia) while maintaining support for languages that are critical for low-latency data acquisition and processing (e.g. C and C++).Approach.To address these needs, we introduce the Backend for Realtime Asynchronous Neural Decoding (BRAND). BRAND comprises Linux processes, termednodes, which communicate with each other in agraphvia streams of data. Its asynchronous design allows for acquisition, control, and analysis to be executed in parallel on streams of data that may operate at different timescales. BRAND uses Redis, an in-memory database, to send data between nodes, which enables fast inter-process communication and supports 54 different programming languages. Thus, developers can easily deploy existing ANN models in BRAND with minimal implementation changes.Main results.In our tests, BRAND achieved <600 microsecond latency between processes when sending large quantities of data (1024 channels of 30 kHz neural data in 1 ms chunks). BRAND runs a brain-computer interface with a recurrent neural network (RNN) decoder with less than 8 ms of latency from neural data input to decoder prediction. In a real-world demonstration of the system, participant T11 in the BrainGate2 clinical trial (ClinicalTrials.gov Identifier: NCT00912041) performed a standard cursor control task, in which 30 kHz signal processing, RNN decoding, task control, and graphics were all executed in BRAND. This system also supports real-time inference with complex latent variable models like Latent Factor Analysis via Dynamical Systems.Significance.By providing a framework that is fast, modular, and language-agnostic, BRAND lowers the barriers to integrating the latest tools in neuroscience and machine learning into closed-loop experiments.

[Ethical considerations for medical applications of implantable brain-computer interfaces]

Implantable brain-computer interfaces (BCIs) have potentially important clinical applications due to the high spatial resolution and signal-to-noise ratio of electrodes that are closer to or implanted in the cerebral cortex. However, the surgery and electrodes of implantable BCIs carry safety risks of brain tissue damage, and their medical applications face ethical challenges, with little literature to date systematically considering ethical norms for the medical applications of implantable BCIs. In order to promote the clinical translation of this type of BCI, we considered the ethics of practice for the medical application of implantable BCIs, including: reducing the risk of brain tissue damage from implantable BCI surgery and electrodes, providing patients with customized and personalized implantable BCI treatments, ensuring multidisciplinary collaboration in the clinical application of implantable BCIs, and the responsible use of implantable BCIs, among others. It is expected that this article will provide thoughts and references for the research and development of ethics of the medical application of implantable BCI.

Identifying distinct neural features between the initial and corrective phases of precise reaching using AutoLFADS

Many initial movements require subsequent corrective movements, but how motor cortex transitions to make corrections and how similar the encoding is to initial movements is unclear. In our study, we explored how the brain's motor cortex signals both initial and corrective movements during a precision reaching task. We recorded a large population of neurons from two male rhesus macaques across multiple sessions to examine the neural firing rates during not only initial movements but also subsequent corrective movements. AutoLFADS, an auto-encoder-based deep-learning model, was applied to provide a clearer picture of neurons' activity on individual corrective movements across sessions. Decoding of reach velocity generalized poorly from initial to corrective submovements. Unlike initial movements, it was challenging to predict the velocity of corrective movements using traditional linear methods in a single, global neural space. We identified several locations in the neural space where corrective submovements originated after the initial reaches, signifying firing rates different than the baseline before initial movements. To improve corrective movement decoding, we demonstrate that a state-dependent decoder incorporating the population firing rates at the initiation of correction improved performance, highlighting the diverse neural features of corrective movements. In summary, we show neural differences between initial and corrective submovements and how the neural activity encodes specific combinations of velocity and position. These findings are inconsistent with assumptions that neural correlations with kinematic features are global and independent, emphasizing that traditional methods often fall short in describing these diverse neural processes for online corrective movements.

Significance Statement

We analyzed submovement neural population dynamics during precision reaching. Using an auto- encoder-based deep-learning model, AutoLFADS, we examined neural activity on a single-trial basis. Our study shows distinct neural dynamics between initial and corrective submovements. We demonstrate the existence of unique neural features within each submovement class that encode complex combinations of position and reach direction. Our study also highlights the benefit of state-specific decoding strategies, which consider the neural firing rates at the onset of any given submovement, when decoding complex motor tasks such as corrective submovements.

Neuromorphic hardware for somatosensory neuroprostheses

In individuals with sensory-motor impairments, missing limb functions can be restored using neuroprosthetic devices that directly interface with the nervous system. However, restoring the natural tactile experience through electrical neural stimulation requires complex encoding strategies. Indeed, they are presently limited in effectively conveying or restoring tactile sensations by bandwidth constraints. Neuromorphic technology, which mimics the natural behavior of neurons and synapses, holds promise for replicating the encoding of natural touch, potentially informing neurostimulation design. In this perspective, we propose that incorporating neuromorphic technologies into neuroprostheses could be an effective approach for developing more natural human-machine interfaces, potentially leading to advancements in device performance, acceptability, and embeddability. We also highlight ongoing challenges and the required actions to facilitate the future integration of these advanced technologies.

Conversion of a medical implant into a versatile computer-brain interface

BACKGROUND

Information transmission into the human nervous system is the basis for a variety of prosthetic applications. Spinal cord stimulation (SCS) systems are widely available, have a well documented safety record, can be implanted minimally invasively, and are known to stimulate afferent pathways. Nonetheless, SCS devices are not yet used for computer-brain-interfacing applications.

OBJECTIVE

Here we aimed to establish computer-to-brain communication via medical SCS implants in a group of 20 individuals who had been operated for the treatment of chronic neuropathic pain.

METHODS

In the initial phase, we conducted interface calibration with the aim of determining personalized stimulation settings that yielded distinct and reproducible sensations. These settings were subsequently utilized to generate inputs for a range of behavioral tasks. We evaluated the required calibration time, task training duration, and the subsequent performance in each task.

RESULTS

We could establish a stable spinal computer-brain interface in 18 of the 20 participants. Each of the 18 then performed one or more of the following tasks: A rhythm-discrimination task (n = 13), a Morse-decoding task (n = 3), and/or two different balance/body-posture tasks (n = 18; n = 5). The median calibration time was 79 min. The median training time for learning to use the interface in a subsequent task was 1:40 min. In each task, every participant demonstrated successful performance, surpassing chance levels.

CONCLUSION

The results constitute the first proof-of-concept of a general purpose computer-brain interface paradigm that could be deployed on present-day medical SCS platforms.

Somatosensory cortex microstimulation modulates primary motor and ventral premotor cortex neurons with extensive spatial convergence and divergence

Intracortical microstimulation (ICMS) is known to affect distant neurons transynaptically, yet the extent to which ICMS pulses delivered in one cortical area modulate neurons in other cortical areas remains largely unknown. Here we assessed how the individual pulses of multi-channel ICMS trains delivered in the upper extremity representation of the macaque primary somatosensory area (S1) modulate neuron firing in the primary motor cortex (M1) and in the ventral premotor cortex (PMv). S1-ICMS pulses modulated the majority of units recorded both in the M1 upper extremity representation and in PMv, producing more inhibition than excitation. Effects converged on individual neurons in both M1 and PMv from extensive S1 territories. Conversely, effects of ICMS delivered in a small region of S1 diverged to wide territories in both M1 and PMv. The effects of this direct modulation of M1 and PMv neurons produced by multi-electrode S1-ICMS like that used here may need to be taken into account by bidirectional brain-computer interfaces that decode intended movements from neural activity in these cortical motor areas.

Significance Statement

Although ICMS is known to produce effects transynaptically, relatively little is known about how ICMS in one cortical area affects neurons in other cortical areas. We show that the effects of multi-channel ICMS in a small patch of S1 diverge to affect neurons distributed widely in both M1 and PMv, and conversely, individual neurons in each of these areas can be affected by ICMS converging from much of the S1 upper extremity representation. Such direct effects of ICMS may complicate the decoding of motor intent from M1 or PMv when artificial sensation is delivered via S1-ICMS in bidirectional brain-computer interfaces.

Plug-and-Play Stability for Intracortical Brain-Computer Interfaces: A One-Year Demonstration of Seamless Brain-to-Text Communication

Intracortical brain-computer interfaces (iBCIs) have shown promise for restoring rapid communication to people with neurological disorders such as amyotrophic lateral sclerosis (ALS). However, to maintain high performance over time, iBCIs typically need frequent recalibration to combat changes in the neural recordings that accrue over days. This requires iBCI users to stop using the iBCI and engage in supervised data collection, making the iBCI system hard to use. In this paper, we propose a method that enables self-recalibration of communication iBCIs without interrupting the user. Our method leverages large language models (LMs) to automatically correct errors in iBCI outputs. The self-recalibration process uses these corrected outputs ("pseudo-labels") to continually update the iBCI decoder online. Over a period of more than one year (403 days), we evaluated our Continual Online Recalibration with Pseudo-labels (CORP) framework with one clinical trial participant. CORP achieved a stable decoding accuracy of 93.84% in an online handwriting iBCI task, significantly outperforming other baseline methods. Notably, this is the longest-running iBCI stability demonstration involving a human participant. Our results provide the first evidence for long-term stabilization of a plug-and-play, high-performance communication iBCI, addressing a major barrier for the clinical translation of iBCIs.

Microstimulation of human somatosensory cortex evokes task-dependent, spatially patterned responses in motor cortex

The primary motor (M1) and somatosensory (S1) cortices play critical roles in motor control but the signaling between these structures is poorly understood. To fill this gap, we recorded - in three participants in an ongoing human clinical trial (NCT01894802) for people with paralyzed hands - the responses evoked in the hand and arm representations of M1 during intracortical microstimulation (ICMS) in the hand representation of S1. We found that ICMS of S1 activated some M1 neurons at short, fixed latencies consistent with monosynaptic activation. Additionally, most of the ICMS-evoked responses in M1 were more variable in time, suggesting indirect effects of stimulation. The spatial pattern of M1 activation varied systematically: S1 electrodes that elicited percepts in a finger preferentially activated M1 neurons excited during that finger's movement. Moreover, the indirect effects of S1 ICMS on M1 were context dependent, such that the magnitude and even sign relative to baseline varied across tasks. We tested the implications of these effects for brain-control of a virtual hand, in which ICMS conveyed tactile feedback. While ICMS-evoked activation of M1 disrupted decoder performance, this disruption was minimized using biomimetic stimulation, which emphasizes contact transients at the onset and offset of grasp, and reduces sustained stimulation.

Preserved neural dynamics across animals performing similar behaviour

Animals of the same species exhibit similar behaviours that are advantageously adapted to their body and environment. These behaviours are shaped at the species level by selection pressures over evolutionary timescales. Yet, it remains unclear how these common behavioural adaptations emerge from the idiosyncratic neural circuitry of each individual. The overall organization of neural circuits is preserved across individuals1 because of their common evolutionarily specified developmental programme2-4. Such organization at the circuit level may constrain neural activity5-8, leading to low-dimensional latent dynamics across the neural population9-11. Accordingly, here we suggested that the shared circuit-level constraints within a species would lead to suitably preserved latent dynamics across individuals. We analysed recordings of neural populations from monkey and mouse motor cortex to demonstrate that neural dynamics in individuals from the same species are surprisingly preserved when they perform similar behaviour. Neural population dynamics were also preserved when animals consciously planned future movements without overt behaviour12 and enabled the decoding of planned and ongoing movement across different individuals. Furthermore, we found that preserved neural dynamics extend beyond cortical regions to the dorsal striatum, an evolutionarily older structure13,14. Finally, we used neural network models to demonstrate that behavioural similarity is necessary but not sufficient for this preservation. We posit that these emergent dynamics result from evolutionary constraints on brain development and thus reflect fundamental properties of the neural basis of behaviour.

Microfluidic-Assembled Covalent Organic Frameworks@Ti3 C2 Tx MXene Vertical Fibers for High-Performance Electrochemical Supercapacitors

The delicate design of innovative and sophisticated fibers with vertical porous skeleton and eminent electrochemical activity to generate directional ionic pathways and good faradic charge accessibility is pivotal but challenging for realizing high-performance fiber-shaped supercapacitors (FSCs). Here, hierarchically ordered hybrid fiber combined vertical-aligned and conductive Ti3 C2 Tx MXene (VA-Ti3 C2 Tx ) with interstratified electroactive covalent organic frameworks LZU1 (COF-LZU1) by one-step microfluidic synthesis is developed. Due to the incorporation of vertical channels, abundant redox active sites and large accessible surface area throughout the electrode, the VA-Ti3 C2 Tx @COF-LZU1 fibers express exceptional gravimetric capacitance of 787 F g-1 in a three-electrode system. Additionally, the solid-state asymmetric FSCs deliver a prominent energy density of 27 Wh kg-1 , capacitance of 398 F g-1 and cycling life of 20 000 cycles. The key to high energy storage ability originates from the decreased ions adsorption energy and ameliorative charge density distribution in vertically aligned and active hybrid fiber, accelerating ions transportation/accommodation and interfacial electrons transfer. Benefiting from excellent electrochemical performance, the FSCs offer sufficient energy supply to power watches, flags, and digital display tubes as well as be integrated with sensors to detect pulse signals, which opens a promising route for architecting advanced fiber toward the carbon neutrality market beyond energy-storage technology.

Neural Data Transformer 2: Multi-context Pretraining for Neural Spiking Activity

The neural population spiking activity recorded by intracortical brain-computer interfaces (iBCIs) contain rich structure. Current models of such spiking activity are largely prepared for individual experimental contexts, restricting data volume to that collectable within a single session and limiting the effectiveness of deep neural networks (DNNs). The purported challenge in aggregating neural spiking data is the pervasiveness of context-dependent shifts in the neural data distributions. However, large scale unsupervised pretraining by nature spans heterogeneous data, and has proven to be a fundamental recipe for successful representation learning across deep learning. We thus develop Neural Data Transformer 2 (NDT2), a spatiotemporal Transformer for neural spiking activity, and demonstrate that pretraining can leverage motor BCI datasets that span sessions, subjects, and experimental tasks. NDT2 enables rapid adaptation to novel contexts in downstream decoding tasks and opens the path to deployment of pretrained DNNs for iBCI control. Code: https://github.com/joel99/context_general_bci.

BRAND: A platform for closed-loop experiments with deep network models

Artificial neural networks (ANNs) are state-of-the-art tools for modeling and decoding neural activity, but deploying them in closed-loop experiments with tight timing constraints is challenging due to their limited support in existing real-time frameworks. Researchers need a platform that fully supports high-level languages for running ANNs (e.g., Python and Julia) while maintaining support for languages that are critical for low-latency data acquisition and processing (e.g., C and C++). To address these needs, we introduce the Backend for Realtime Asynchronous Neural Decoding (BRAND). BRAND comprises Linux processes, termed nodes , which communicate with each other in a graph via streams of data. Its asynchronous design allows for acquisition, control, and analysis to be executed in parallel on streams of data that may operate at different timescales. BRAND uses Redis to send data between nodes, which enables fast inter-process communication and supports 54 different programming languages. Thus, developers can easily deploy existing ANN models in BRAND with minimal implementation changes. In our tests, BRAND achieved <600 microsecond latency between processes when sending large quantities of data (1024 channels of 30 kHz neural data in 1-millisecond chunks). BRAND runs a brain-computer interface with a recurrent neural network (RNN) decoder with less than 8 milliseconds of latency from neural data input to decoder prediction. In a real-world demonstration of the system, participant T11 in the BrainGate2 clinical trial performed a standard cursor control task, in which 30 kHz signal processing, RNN decoding, task control, and graphics were all executed in BRAND. This system also supports real-time inference with complex latent variable models like Latent Factor Analysis via Dynamical Systems. By providing a framework that is fast, modular, and language-agnostic, BRAND lowers the barriers to integrating the latest tools in neuroscience and machine learning into closed-loop experiments.

Biomimetic multi-channel microstimulation of somatosensory cortex conveys high resolution force feedback for bionic hands

Manual interactions with objects are supported by tactile signals from the hand. This tactile feedback can be restored in brain-controlled bionic hands via intracortical microstimulation (ICMS) of somatosensory cortex (S1). In ICMS-based tactile feedback, contact force can be signaled by modulating the stimulation intensity based on the output of force sensors on the bionic hand, which in turn modulates the perceived magnitude of the sensation. In the present study, we gauged the dynamic range and precision of ICMS-based force feedback in three human participants implanted with arrays of microelectrodes in S1. To this end, we measured the increases in sensation magnitude resulting from increases in ICMS amplitude and participant's ability to distinguish between different intensity levels. We then assessed whether we could improve the fidelity of this feedback by implementing "biomimetic" ICMS-trains, designed to evoke patterns of neuronal activity that more closely mimic those in natural touch, and by delivering ICMS through multiple channels at once. We found that multi-channel biomimetic ICMS gives rise to stronger and more distinguishable sensations than does its single-channel counterpart. Finally, we implemented biomimetic multi-channel feedback in a bionic hand and had the participant perform a compliance discrimination task. We found that biomimetic multi-channel tactile feedback yielded improved discrimination over its single-channel linear counterpart. We conclude that multi-channel biomimetic ICMS conveys finely graded force feedback that more closely approximates the sensitivity conferred by natural touch.

Microstimulation of sensory cortex engages natural sensory representations

Cortical activity patterns occupy a small subset of possible network states. If this is due to intrinsic network properties, microstimulation of sensory cortex should evoke activity patterns resembling those observed during natural sensory input. Here, we use optical microstimulation of virally transfected layer 2/3 pyramidal neurons in the mouse primary vibrissal somatosensory cortex to compare artificially evoked activity with natural activity evoked by whisker touch and movement ("whisking"). We find that photostimulation engages touch- but not whisking-responsive neurons more than expected by chance. Neurons that respond to photostimulation and touch or to touch alone exhibit higher spontaneous pairwise correlations than purely photoresponsive neurons. Exposure to several days of simultaneous touch and optogenetic stimulation raises both overlap and spontaneous activity correlations among touch and photoresponsive neurons. We thus find that cortical microstimulation engages existing cortical representations and that repeated co-presentation of natural and artificial stimulation enhances this effect.

[Ethics considerations on brain-computer interface technology]

The development and potential application of brain-computer interface (BCI) technology is closely related to the human brain, so that the ethical regulation of BCI has become an important issue attracting the consideration of society. Existing literatures have discussed the ethical norms of BCI technology from the perspectives of non-BCI developers and scientific ethics, while few discussions have been launched from the perspective of BCI developers. Therefore, there is a great need to study and discuss the ethical norms of BCI technology from the perspective of BCI developers. In this paper, we present the user-centered and non-harmful BCI technology ethics, and then discuss and look forward on them. This paper argues that human beings can cope with the ethical issues arising from BCI technology, and as BCI technology develops, its ethical norms will be improved continuously. It is expected that this paper can provide thoughts and references for the formulation of ethical norms related to BCI technology.

S1 represents multisensory contexts and somatotopic locations within and outside the bounds of the cortical homunculus

Recent literature suggests that tactile events are represented in the primary somatosensory cortex (S1) beyond its long-established topography; in addition, the extent to which S1 is modulated by vision remains unclear. To better characterize S1, human electrophysiological data were recorded during touches to the forearm or finger. Conditions included visually observed physical touches, physical touches without vision, and visual touches without physical contact. Two major findings emerge from this dataset. First, vision strongly modulates S1 area 1, but only if there is a physical element to the touch, suggesting that passive touch observation is insufficient to elicit neural responses. Second, despite recording in a putative arm area of S1, neural activity represents both arm and finger stimuli during physical touches. Arm touches are encoded more strongly and specifically, supporting the idea that S1 encodes tactile events primarily through its topographic organization but also more generally, encompassing other areas of the body.

Regressive changes in sizes of somatosensory cuneate nucleus after sensory loss in primates

Neurons in the early stages of processing sensory information suffer transneuronal atrophy when deprived of their activating inputs. For over 40 y, members of our laboratory have studied the reorganization of the somatosensory cortex during and after recovering from different types of sensory loss. Here, we took advantage of the preserved histological material from these studies of the cortical effects of sensory loss to evaluate the histological consequences in the cuneate nucleus of the lower brainstem and the adjoining spinal cord. The neurons in the cuneate nucleus are activated by touch on the hand and arm, and relay this activation to the contralateral thalamus, and from the thalamus to the primary somatosensory cortex. Neurons deprived of activating inputs tend to shrink and sometimes die. We considered the effects of differences in species, type and extent of sensory loss, recovery time after injury, and age at the time of injury on the histology of the cuneate nucleus. The results indicate that all injuries that deprived part or all of the cuneate nucleus of sensory activation result in some atrophy of neurons as reflected by a decrease in nucleus size. The extent of the atrophy is greater with greater sensory loss and with longer recovery times. Based on supporting research, atrophy appears to involve a reduction in neuron size and neuropil, with little or no neuron loss. Thus, the potential exists for restoring the hand to cortex pathway with brain-machine interfaces, for bionic prosthetics, or biologically with hand replacement surgery.

Proprioception: A New Era Set in Motion by Emerging Genetic and Bionic Strategies?

The generation of an internal body model and its continuous update is essential in sensorimotor control. Although known to rely on proprioceptive sensory feedback, the underlying mechanism that transforms this sensory feedback into a dynamic body percept remains poorly understood. However, advances in the development of genetic tools for proprioceptive circuit elements, including the sensory receptors, are beginning to offer new and unprecedented leverage to dissect the central pathways responsible for proprioceptive encoding. Simultaneously, new data derived through emerging bionic neural machine-interface technologies reveal clues regarding the relative importance of kinesthetic sensory feedback and insights into the functional proprioceptive substrates that underlie natural motor behaviors.

PHYSICAL FITNESS IMPROVEMENTS IN FEMALE VOLLEYBALL PLAYERS

ABSTRACT Introduction: Volleyball is a competitive sport demanding the athlete's excellent physical fitness because the execution of its movements requires flexibility and agility. Objective: Analyze the effects of a specific training protocol on physical fitness at a competitive level on volleyball players. Methods: This paper selects 12 female volleyball players as volunteers for research subjects. The athletes’ agility quality was examined for intellectual efficiency, psychological susceptibility, and flexibility, among other indicators. The mathematical statistical analysis program SPSS13.0 was used to analyze and process female volleyball players’ sensitivity and characteristics of fitness quality. Results: The responsiveness of Chinese female volleyball players is mainly reflected in athletes’ intellectual agility and kinetic adaptations. The characteristics of technology for protection are practical, diverse, and economical. Transmission technology is characterized by high speed and quick thinking. The sensitivity and quality of the volleyball players are high. Conclusion: The sensitivity of volleyball players is produced in response to changes in the sports court environment. The use of various tactics and auxiliary equipment in volleyball will cause changes in the court situation. The quality of sensitivity is critical in the physical training of volleyball players. Level of evidence II; Therapeutic studies - investigation of treatment outcomes.

Learning in a closed-loop brain-machine interface with distributed optogenetic cortical feedback

Objective.Distributed microstimulations at the cortical surface can efficiently deliver feedback to a subject during the manipulation of a prosthesis through a brain-machine interface (BMI). Such feedback can convey vast amounts of information to the prosthesis user and may be key to obtain an accurate control and embodiment of the prosthesis. However, so far little is known of the physiological constraints on the decoding of such patterns. Here, we aimed to test a rotary optogenetic feedback that was designed to encode efficiently the 360° movements of the robotic actuators used in prosthetics. We sought to assess its use by mice that controlled a prosthesis joint through a closed-loop BMI.Approach.We tested the ability of mice to optimize the trajectory of a virtual prosthesis joint in order to solve a rewarded reaching task. They could control the speed of the joint by modulating the activity of individual neurons in the primary motor cortex. During the task, the patterned optogenetic stimulation projected on the primary somatosensory cortex continuously delivered information to the mouse about the position of the joint.Main results.We showed that mice are able to exploit the continuous, rotating cortical feedback in the active behaving context of the task. Mice achieved better control than in the absence of feedback by detecting reward opportunities more often, and also by moving the joint faster towards the reward angular zone, and by maintaining it longer in the reward zone. Mice controlling acceleration rather than speed of the joint failed to improve motor control.Significance.These findings suggest that in the context of a closed-loop BMI, distributed cortical feedback with optimized shapes and topology can be exploited to control movement. Our study has direct applications on the closed-loop control of rotary joints that are frequently encountered in robotic prostheses.

Clinical neuroscience and neurotechnology: An amazing symbiosis

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

Getting a grasp on BMIs: Decoding prehension and speech signals

Wandelt et al. (2022) show that different grasps can be decoded from neural activity in the human supramarginal gyrus (SMG), ventral premotor cortex, and somatosensory cortex during motor imagery and speech, highlighting the attractiveness of higher-level areas such as the SMG for brain-machine interface applications.

Behaviorist approaches to investigating memory and learning: A primer for synthetic biology and bioengineering

The fields of developmental biology, biomedicine, and artificial life are being revolutionized by advances in synthetic morphology. The next phase of synthetic biology and bioengineering is resulting in the construction of novel organisms (biobots), which exhibit not only morphogenesis and physiology but functional behavior. It is now essential to begin to characterize the behavioral capacity of novel living constructs in terms of their ability to make decisions, form memories, learn from experience, and anticipate future stimuli. These synthetic organisms are highly diverse, and often do not resemble familiar model systems used in behavioral science. Thus, they represent an important context in which to begin to unify and standardize vocabulary and techniques across developmental biology, behavioral ecology, and neuroscience. To facilitate the study of behavior in novel living systems, we present a primer on techniques from the behaviorist tradition that can be used to probe the functions of any organism – natural, chimeric, or synthetic – regardless of the details of their construction or origin. These techniques provide a rich toolkit for advancing the fields of synthetic bioengineering, evolutionary developmental biology, basal cognition, exobiology, and robotics.