Biology subjects

Michaels, J. A.

Publications and source records attributed to Michaels, J. A..

15 recordsLinked to original sources

bioRxiv2026

Reward-driven emergence of auditory pattern encoding in the primate motor system

The ability to anticipate rhythmic patterns is fundamental to human experience, enabling music appreciation, speech comprehension, and dancing in sync to music. How the brain learns to use acoustic information to guide motor behavior remains a key question, with its neural underpinnings and evolutionary origins debated, especially in non-human primates. Reward contingencies improve monkeys ability to synchronize with musical beats with near-human accuracy. However, humans and certain birds spontaneously synchronize with musical beats. Behavioural work in monkeys suggests that the brains reward system can encode rhythmic patterns if sufficiently reinforced, known as the intrinsic reward hypothesis. To investigate the neural mechanism of how the reward gates rhythmic pattern learning, we recorded neural activity from the globus pallidus interna (GPi) of a rhesus macaque during passive listening to auditory patterns. We found that GPis ability to encode auditory patterns in a monkey is shaped by reward association. Specifically, during passive listening (Experiment 1), with reward only at the end of each trial, GPi showed at most a weak response to the auditory tone patterns that was not evident in single neurons or population responses, ruling out spontaneous entrainment of motor activity to auditory patterns. When each tone was simultaneously accompanied by a liquid reward (Experiment 2), GPi showed clear responses to the tone patterns. To test if the tone responses persisted, the reward contingency was subsequently removed (Experiment 3). Strikingly, although the reward was delivered only at the trials end as in Experiment 1, GPi neurons continued to track the occurrence of the tones. Besides GPi, we also recorded single-neuron activity across the primary somatosensory (S1), primary motor (M1), dorsal premotor (PMd), supplementary motor (SMA), pre-supplementary motor (preSMA) cortices, and the medial geniculate body (MGB) of the same animal in experiments 1 and 2. We found that during passive listening (Experiment 1) with a reward only at the end of each trial, primarily the MGB, not motor areas, responded to the auditory tone patterns, ruling out spontaneous entrainment of motor activity to auditory patterns across the primate motor system. When each tone was simultaneously accompanied by a liquid reward (Experiment 2), all areas except S1 showed clear responses to the tone patterns. These findings suggest that while basic sensory processing of tone patterns occurs in the MGB, the GPi exhibits reward-driven encoding of auditory patterns that persists after reward withdrawal. This reorganization of neural responsiveness shows that the capacity for auditory association can be elicited in the motor system through appropriate motivational contexts.

neuroscience↗

bioRxiv2026

Sensory expectations and prediction error during feedback control in the human brain

External disturbances to the body are better counteracted when their nature can be predicted in advance. Here, we investigated the neural mechanisms through which probabilistic predictions shape feedback responses using functional magnetic resonance imaging (fMRI) in humans. We show that, prior to a mechanical perturbation applied to a finger, the primary motor (M1) and somatosensory (S1) cortices receive a signal that linearly encodes the expected sensory input. When perturbations reach these areas, expectations are combined with the sensory input through a simple additive mechanism, yielding motor commands that reflect a weighted sum of the two signals. At the same time, M1 and S1 receive a prediction error signal encoding the difference between expectations and actual sensory input. This signal is visible in fMRI data in humans and in the local field potentials in non-human primates, but not in M1 or S1 spiking activity.

neuroscience↗

bioRxiv2026

High performance sorting of motor unit action potentials with EMUsort

Understanding how neural signals control muscle activity during behavior is a key challenge in motor neuroscience. To this end, recent advances in intramuscular multielectrode arrays have enabled high-quality multichannel recordings of many motor unit action potentials (MUAPs) in freely moving subjects. However, identifying individual MUAP events within multichannel recordings is a significant challenge for existing spike sorting methods, which are typically optimized for identifying action potentials from neurons in the brain. To overcome this challenge, we developed the Enhanced Motor Unit sorter (EMUsort), an extension of Kilosort4 (KS4) that achieves high-performance MUAP spike sorting. We applied EMUsort to high-resolution intramuscular recordings from rat forelimb during locomotion and monkey forelimb during a reaching task. EMUsort improves upon prior methods by addressing key challenges encountered with MUAP datasets, including: 1) long time delays across electrodes due to propagation along muscle fibers, 2) more complex waveform shapes compared to neuronal action potentials, and 3) a high degree of MUAP overlap due to cumulative motor unit recruitment. We compared EMUsort to existing spike sorting methods quantitatively using simulated datasets that closely emulated the rat and monkey datasets we recorded. EMUsort provided median error rate reductions of 67.5% and 49.9% during periods of high motor unit activation for the rat and monkey datasets, respectively. In sum, EMUsort provides a substantial improvement to MUAP spike sorter accuracy, especially during regions of high MUAP overlap, in an easy-to-use software package.

neuroscience↗

bioRxiv2025

Compositional neural dynamics during reaching

The complex mechanics of the arm make the neural control of reaching inherently posture dependent. Because previous reaching studies confound reach direction with final posture, it remains unknown how neural population dynamics in the motor cortex account for arm posture. Here we address this gap with high-density neural recordings and a reaching task in which the same targets serve as start points on some trials and end points on others. We show that neural population dynamics in monkey primary motor cortex and dorsal premotor cortex exhibit a compositional structure with three components that enable posture-dependent control: first, a posture subspace containing fixed points visited whenever the arm is in a specific posture; second, rotational dynamics that transition between these fixed points, systematically organized so that similar rotations produce similar movements while continuously updating the posture representation; third, a condition-independent shift dimension that tracks trial progression across all movements. This compositional structure advances the population-level account of how motor cortical dynamics support skilled reaching.

neuroscience↗

bioRxiv2025

ATHENA: Automatically Tracking Hands Expertly with No Annotations

Studying naturalistic hand behaviours is challenging due to the limitations of conventional marker-based motion capture, which can be costly, time-consuming, and encumber participants. While markerless pose estimation exists - an accurate, off-the-shelf solution validated for hand-object manipulation is needed. We present ATHENA (Automatically Tracking Hands Expertly with No Annotations), an open-source, Python-based toolbox for 3D markerless hand tracking. To validate ATHENA, we concurrently recorded hand kinematics using ATHENA and an industry-standard optoelectronic marker-based system (OptiTrack). Participants performed unimanual, bimanual, and naturalistic object manipulation and we compared common kinematic variables like grip aperture, wrist velocity, index metacarpophalangeal flexion, and bimanual span. Our results demonstrated high spatiotemporal agreement between ATHENA and OptiTrack. This was evidenced by extremely high matches (R2 > 0.90 across the majority of tasks) and low root mean square differences (< 1 cm for grip aperture, < 4 cm/s for wrist velocity, and < 5-10{degrees} for index metacarpophalangeal flexion). ATHENA reliably preserved trial-to-trial variability in kinematics, offering identical scientific conclusions to marker-based approaches, but with significantly reduced financial and time costs and no participant encumbrance. In conclusion, ATHENA is an accurate, automated, and easy-to-use platform for 3D markerless hand tracking that enables more ecologically valid motor control and learning studies of naturalistic hand behaviours, enhancing our understanding of human dexterity.

neuroscience↗

bioRxiv2025

A Context-Free Model of Savings in Motor Learning

Learning to adapt voluntary movements to an external perturbation, whether mechanical or visual, is faster during a second encounter than during the first. The mechanisms underlying this phenomenon, known as savings, remain unclear. Recent studies propose that the high dimensionality of neural control enables the retention of learning traces that may facilitate savings. To test this idea we used MotorNet, a framework for training recurrent neural networks (RNNs) to control biomechanical models of the human upper limb. RNNs were trained to perform reaching movements with a velocity-dependent force field (FF) and without (NF) in the sequence NF1 (baseline), FF1 (adaptation), NF2 (washout), and FF2 (re-adaptation). RNNs showed behavioural signatures of savings in the absence of any explicit contextual input signalling the presence or absence of the FF. Savings was more robust in RNNs with larger numbers of units. We identified a component of RNN activity associated with savings--a shift in preparatory activity that persisted even after washout. Displacing this preparatory activity in the direction of the shift enhanced savings, whereas perturbations in the opposite direction reduced or eliminated savings. These findings suggest a potential neural basis for motor memory retention underlying savings that is reliant on the high dimensionality of neural circuits for control, and is independent of cognitive or strategic learning.

neuroscience↗

bioRxiv2025

A paradigm to study the learning of muscle activity patterns outside of the natural repertoire

The acquisition of novel muscle activity patterns is a key aspect of motor skill learning which can be seen at play, for example, when beginner musicians learn new guitar or piano chords. To study this process, here we introduce a new paradigm that requires quick and synchronous flexion and extension of multiple fingers. First, participants practiced all the 242 possible combinations of isometric finger flexion and extension around the metacarpophalangeal joint (i.e., chords). We found that some chords were initially extremely challenging, but participants could eventually achieve them with practice, showing that the initial difficulty did not reflect hard biomechanical constraints imposed by the interaction of tendons and ligaments. In a second experiment we found that chord learning was largely chord-specific and did not generalize to untrained chords. Finally, we explored which factors make some chords more difficult than others. Difficulty was well predicted by the muscle activity pattern required by that chord. Interestingly, difficulty related to how similar chords muscle activity patterns are to the muscle activity patterns required by everyday hand use, and to the overall size of the muscle activity. Together, our results suggest that the new paradigm introduced in this work may provide a valuable tool to study the neural processes underlying the acquisition of novel muscle activity patterns in the human motor system.

neuroscience↗

bioRxiv2025

Computation-through-Dynamics Benchmark: Simulated datasets and quality metrics for dynamical models of neural activity

A primary goal of systems neuroscience is to discover how ensembles of neurons transform inputs into goal-directed behavior, a process known as neural computation. A powerful framework for understanding neural computation uses neural dynamics - the rules that govern how neural activity evolves over time - to explain how goal-directed input-output transformations occur. As dynamical rules are not directly observable, we need computational models that can infer neural dynamics from recorded neural activity. We typically validate such models using synthetic datasets with known ground-truth dynamics, but unfortunately existing synthetic datasets dont reflect fundamental features of neural computation and may therefore be poor proxies for neural systems. Further, the field lacks validated metrics for quantifying the accuracy of the dynamics inferred by models. The Computation-through-Dynamics Toolkit (CtDToolkit) addresses these critical gaps by providing: 1) synthetic datasets that reflect computational properties of biological neural circuits, 2) interpretable metrics for quantifying model performance, and 3) a standardized pipeline for training and evaluating models with or without known external inputs. In this manuscript, we demonstrate how CtDToolkit can help guide the development, tuning, and troubleshooting of neural dynamics models. In summary, CtDToolkit provides a necessary framework for model developers to better understand and characterize neural computation through the lens of dynamics. Author SummaryUnderstanding how the brain works requires interpretable accounts of how populations of neurons process information to produce behavior. One powerful approach is to study "neural dynamics", the patterns of how neural activity evolves over time. Scientists develop computational models to infer these dynamics from neural recordings, but it has been challenging to know when the inferred dynamics are trustworthy. Existing datasets often lack key features of biological neural circuits, and current performance metrics can provide an incomplete picture of model quality. We developed the Computation-through-Dynamics Toolkit (CtDToolkit) to solve these problems. Our toolkit provides three key resources: biologically motivated synthetic datasets, improved metrics that provide more holistic accounts of model performance, and a standardized workflow for training and evaluating models. We hope that CtDToolkit enables researchers to rigorously test, improve, and troubleshoot their models before applying them to real brain data. This work establishes a crucial foundation for developing better methods to understand neural computation, ultimately advancing our ability to decode how the brain transforms sensory information into thought and action.

neuroscience↗

bioRxiv2024

Sensory expectations shape neural population dynamics in motor circuits

The neural basis of movement preparation has been extensively studied during self-initiated actions where motor cortical activity during preparation shows a lawful relationship to the parameters of the subsequent action1,2. However, movements are regularly triggered and constantly corrected based on sensory inputs caused by disturbances to the body or environment. Since such disturbances are often predictable and since preparing for disturbances would make movements better, we hypothesized that expectations about sensory inputs also influence preparatory activity in motor circuits. Here we show that when humans and monkeys are probabilistically cued about the direction of a future mechanical perturbation, they incorporate sensory expectations into their movement preparation and improve their corrective responses. Using high-density neural recordings, we establish that sensory expectations are widespread across the brain, including the motor cortical areas involved in preparing self-initiated actions. The geometry of these preparatory signals in the neural population state is simple, scaling directly with the probability of each perturbation direction. After perturbation onset, a condition-independent perturbation signal shifts the neural state leading to rapid responses that initially reflect sensory expectations. Based on neural networks coupled to a biomechanical model of the arm3, we show that this neural geometry emerges through training, but only when the incoming sensory information indicating perturbation direction coincides with - or is preceded by - a condition-independent signal indicating that a perturbation has occurred. Thus, motor circuit dynamics are shaped by future sensory inputs, providing clear empirical support for the idea that movement is governed by the sophisticated manipulation of sensory feedback4.

neuroscience↗

bioRxiv2023

Accurate neuroprosthetic control through latent state transition training

Brain-computer interfaces (BCIs) have the potential to restore hand movement for people with paralysis, but current devices still lack the fine control required to interact with objects of daily living. Following understanding of cortical activity during arm reaches, hand BCI studies have focused on velocity control. However, mounting evidence suggests that posture, and not velocity, dominates in hand-related areas during natural movement. To explore whether this signal can causally control a prosthesis, we developed a novel BCI training paradigm centered on the reproduction of hand posture transitions. Macaque monkeys trained with the protocol were able to control a multi-dimensional hand prosthesis at high-accuracy, including execution of the very intricate precision grip. Subsequent analysis revealed that the posture signal in the target grasping areas was a major contributor to control. Population activity exhibited pattern separation and dimensionality increases driven by posture kinematics, and simulations with a grasping circuit model demonstrated the generalizability of our approach. We present for the first time neural posture control of a multi-dimensional hand prosthesis, opening the door for future devices to leverage this additional information channel.

neuroscience↗

bioRxiv2023

Myomatrix arrays for high-definition muscle recording

Neurons coordinate their activity to produce an astonishing variety of motor behaviors. Our present understanding of motor control has grown rapidly thanks to new methods for recording and analyzing populations of many individual neurons over time. In contrast, current methods for recording the nervous systems actual motor output - the activation of muscle fibers by motor neurons - typically cannot detect the individual electrical events produced by muscle fibers during natural behaviors and scale poorly across species and muscle groups. Here we present a novel class of electrode devices ("Myomatrix arrays") that record muscle activity at unprecedented resolution across muscles and behaviors. High-density, flexible electrode arrays allow for stable recordings from the muscle fibers activated by a single motor neuron, called a "motor unit", during natural behaviors in many species, including mice, rats, primates, songbirds, frogs, and insects. This technology therefore allows the nervous systems motor output to be monitored in unprecedented detail during complex behaviors across species and muscle morphologies. We anticipate that this technology will allow rapid advances in understanding the neural control of behavior and in identifying pathologies of the motor system.

neuroscience↗

bioRxiv2023

MotorNet: a Python toolbox for controlling differentiable biomechanical effectors with artificial neural networks

Artificial neural networks (ANNs) are a powerful class of computational models for unravelling neural mechanisms of brain function. However, for neural control of movement, they currently must be integrated with software simulating biomechanical effectors, leading to limiting impracticalities: (1) researchers must rely on two different platforms and (2) biomechanical effectors are not generally differentiable, constraining researchers to reinforcement learning algorithms despite the existence and potential biological relevance of faster training methods. To address these limitations, we developed MotorNet, an open-source Python toolbox for creating arbitrarily complex, differentiable, and biomechanically realistic effectors that can be trained on user-defined motor tasks using ANNs. MotorNet is designed to meet several goals: ease of installation, ease of use, a high-level user-friendly API, and a modular architecture to allow for flexibility in model building. MotorNet requires no dependencies outside Python, making it easy to get started with. For instance, it allows training ANNs on typically used motor control models such as a two joint, six muscle, planar arm within minutes on a typical desktop computer. MotorNet is built on PyTorch and therefore can implement any network architecture that is possible using the PyTorch framework. Consequently, it will immediately benefit from advances in artificial intelligence through PyTorch updates. Finally, it is open source, enabling users to create and share their own improvements, such as new effector and network architectures or custom task designs. MotorNets focus on higher order model and task design will alleviate overhead cost to initiate computational projects for new researchers by providing a standalone, ready-to-go framework, and speed up efforts of established computational teams by enabling a focus on concepts and ideas over implementation.

neuroscience↗

bioRxiv2019

A neural network model of flexible grasp movement generation

One of the primary ways we interact with the world is using our hands. In macaques, the circuit spanning the anterior intraparietal area, the hand area of the ventral premotor cortex, and the primary motor cortex is necessary for transforming visual information into grasping movements. We hypothesized that a recurrent neural network mimicking the multi-area structure of the anatomical circuit and using visual features to generate the required muscle dynamics to grasp objects would explain the neural and computational basis of the grasping circuit. Modular networks with object feature input and sparse inter-module connectivity outperformed other models at explaining neural data and the inter-area relationships present in the biological circuit, despite the absence of neural data during network training. Network dynamics were governed by simple rules, and targeted lesioning of modules produced deficits similar to those observed in lesion studies, providing a potential explanation for how grasping movements are generated.

neuroscience↗

bioRxiv2017

Neural dynamics of variable grasp movement preparation in the macaque fronto-parietal network

Our voluntary grasping actions lie on a continuum between immediate action and waiting for the right moment, depending on the context. Therefore, studying grasping requires investigating how preparation time affects this process. Two macaque monkeys (Macaca mulatta) performed a grasping task with a short instruction followed by an immediate or delayed go cue (0-1300 ms) while we recorded in parallel from neurons in the hand area (F5) of the ventral premotor cortex and the anterior intraparietal area (AIP). Initial population dynamics followed a fixed trajectory in the neural state space unique to each grip type, reflecting unavoidable preparation, then diverged depending on the delay. Although similar types of single unit responses were present in both areas, population activity in AIP stabilized within a unique memory state while F5 activity continued to evolve, tracking subjective anticipation of the go cue. Intriguingly, activity during movement initiation clustered into two trajectory clusters, corresponding to movements that were either as fast as possible or withheld movements, demonstrating a widespread state shift in the fronto-parietal grasping network when movements must be withheld. Our results reveal how dissociation between static and dynamic components of movement preparation as well as differentiation between cortical areas is possible through population level analysis.\n\nSignificance StatementMany of our movements must occur with no warning, while others we can prepare in advance. Yet, its unclear how planning for movements along the spectrum between these two situations differs in the brain. Two macaque monkeys made reach to grasp movements after varying amounts of preparation time while we recorded from premotor and parietal cortex. We found that the initial response to a grasp instruction was specific to the required movement, but not the preparation time, reflecting required processing. However, when more preparation time was given, neural activity achieved unique states that likely related to withholding movements and anticipation of movement, which was more prevalent in premotor cortex, suggesting differing roles of premotor and parietal cortex in grasp planning.

neuroscience↗

bioRxiv2017

Population coding of grasp and laterality-related information in the macaque fronto-parietal network

Preparing and executing grasping movements demands the coordination of sensory information across multiple scales. The position of an object, required hand shape, and which of our hands to extend must all be coordinated in parallel. The network formed by the macaque anterior intraparietal area (AIP) and hand area (F5) of the ventral premotor cortex is essential in the generation of grasping movements. Yet, the role of this circuit in hand selection is unclear. We recorded from 1342 single- and multi-units in AIP and F5 of two macaque monkeys (Macaca mulatta) during a delayed grasping task in which monkeys were instructed by a visual cue to perform power or precision grips on a handle presented in five different orientations with either the left or right hand, as instructed by an auditory tone. In AIP, intended hand use was only weakly represented during preparation, while hand use was robustly present in F5 during preparation. Interestingly, visual-centric handle orientation information dominated AIP, while F5 contained an additional body-centric frame during preparation and movement. Together, our results implicate F5 as a site of visuo-motor transformation and advocate a strong transition between hand-invariant and hand-specific representations in this parieto-frontal circuit.

neuroscience↗