Single Unit publications

How the brain organizes discrete actions into fluid sequences is a central problem in motor neuroscience. Competing models of basal ganglia (BG) function propose that BG neurons either signal sequence boundaries or encode movements across ordinal positions. Prior studies have largely examined fixed sequences with end-of-sequence rewards, leaving open whether such findings generalize to more naturalistic conditions. We trained four rhesus macaques to perform a visuomotor sequence task requiring four or five out-and-back joystick movements to peripheral targets. Sequences were completed under two conditions: a random condition, in which target order varied across trials, and a fixed condition, in which order was predictable and consistent. Rewards were delivered after each movement, dissociating reward timing from sequence completion. We recorded single-unit activity in arm-related regions of the globus pallidus (GP; n = 458) and primary motor cortex (M1; n = 306). Regression analyses revealed that many neurons in both GP and M1 encoded ordinal position within a sequence. Order effects were more frequent in the fixed condition, but were also present during random sequences. We found no evidence for preferential encoding of sequence initiation or termination in overlearned sequences, in contrast to prior studies reporting start/stop signals in basal ganglia. Weak effects appeared under the random condition in one animal pair, but these did not generalize across animals or conditions. Instead, neurons exhibited heterogeneous order-related responses spanning the full sequence. These results demonstrate that GP neurons, like those in M1, encode ordinal position throughout a sequence rather than acting solely as sequence initiators or terminators. This challenges boundary-specific models of BG function and highlights the BG’s broader role in representing serial order during motor sequence production.

The locus coeruleus–norepinephrine (LC-NE) system regulates arousal and awakening; however, it remains unclear whether the LC does this in a global or circuit-specific manner. We hypothesized that sensory-evoked awakenings are predominantly regulated by specific LC-NE efferent pathways. Anatomical, physiological, and functional modularities of LC-NE pathways involving the mouse basal forebrain (BF) and pontine reticular nucleus (PRN) were tested. We found partial anatomical segregation between the LC → PRN and LC → BF circuits. Extracellular NE dynamics in BF and PRN exhibited distinct sound-evoked activation during sleep, including a fast sound-evoked NE peak specific to PRN. Causal optogenetic interrogation of LC efferent pathways, by retro-channelrhodopsin (ChR2) activation or Platynereis dumerilii ciliary opsin (PdCO) silencing of synapses in target regions, revealed a role for early LC → PRN activity in driving arousal and sound-evoked awakenings. Together, our results uncover a role for early LC-NE PRN activity in connecting sensory and arousal pathways and establish LC heterogeneity in regulating arousal.

The ability to recognize speakers by their voice despite acoustical variation plays a substantial role in primate social interactions. Although neurons in the macaque anterior temporal lobe (ATL) show invariance to face viewpoint, whether they also encode abstract representations of caller identity is not known. Here, we demonstrate that neurons in the voice-selective ATL of two macaques support invariant voice identity representations through dynamic population-level coding. These representations minimize neural distances across different vocalizations from the same individual while preserving distinct trajectories for different callers. This structure emerged from the coordinated activity of the broader neuronal ensemble, although a small subset of highly identity-selective neurons carried high identity information. Our findings provide a neural basis for voice identity recognition in primates and highlight the ATL as a key hub for integrating perceptual voice features into higher-level identity representations.

Brain-machine interfaces (BMIs) offer a viable option for restoring function in patients with motor disabilities post-stroke. Most BMI systems rely on signals from the motor cortex (M1), which is often compromised after stroke. The cerebellum, a subcortical structure involved in motor control, remains an underexplored source for neuroprosthetic control. Using chronic electrophysiological recordings in a rat stroke model, we show that cerebellar neural activity can effectively drive BMI control, performing comparably to M1-driven control. We observed this even in animals with motor impairments post-stroke. Simultaneous M1-cerebellum recordings during cerebellar BMI control revealed that cerebellar "direct" neurons driving the interface were influenced by both local cerebellar and distant M1 neurons. While cerebellar influence remained stable, M1's interaction with cerebellar direct neurons shifted from longer to shorter timescales after stroke. These findings highlight that cerebellar direct neural control is possible in the stroke brain and reveal changes in M1-cerebellar network dynamics post-stroke.

The central nervous system controls movement by combining neuromotor modules, known as muscle synergies. Previous studies suggest that spinal premotor interneurons (PreM‐INs) contribute to the encoding of stable muscle synergies for voluntary movement. But descending and sensory inputs also influence motor outputs through the spinal interneuronal network, which may be configured by its inputs to encode different sets of muscle synergies depending on the network state, thereby recruiting different selections of synergies. Here we tested this possibility of state‐dependent synergy encoding by examining the muscle synergies represented by the same upstream spinal interneurons under different activity states induced by various optogenetic stimulation patterns. Lumbosacral spinal units and electromyographic (EMG) activities of hindlimb muscles were simultaneously recorded from anaesthetized Thy1‐ChR2 mice as the spinal cord was stimulated by one or two optic fibres at different intensities. The synergy encoded by each unit was revealed as a ‘muscle field’ derived from spike‐triggered averages of EMG, whereas the entire muscle synergy set was factorized from the EMG. We found that although the muscle synergy set remained stable across stimulation conditions, the muscle fields of the same units were matched to different synergies within the set in different states. Thus the interneurons may flexibly adjust their connectivity with the motoneurons of the muscles as descending and sensory afferents impose different states on the spinal network. State‐dependent encoding of muscle synergies may allow different synergies to be selected for producing stable movement in an ever‐changing workspace environment. image Key points Muscle synergies for locomotion can be represented by spinal interneurons, as revealed by the interneurons’ muscle fields derived from spike‐triggered averages of EMG. The muscle field of a single spinal interneuron may vary under different stimulation conditions, as demonstrated by optogenetic stimulation. Encoding of muscle synergies is dependent on the state of spinal activities, thus facilitating the selection of appropriate synergies in different dynamic environments.

Atypical sensory processing, particularly in the auditory domain, is one of the most common and quality-of-life affecting symptoms seen in autism spectrum disorders (ASD). Fragile X Syndrome (FXS) is a leading inherited cause of ASD and a majority of FXS individuals present with auditory processing alterations. While auditory hypersensitivity is a common phenotype observed in FXS and Fmr1 knockout (KO) rodent models, it is important to consider other auditory coding impairments that could contribute to sound processing difficulties and disrupted language comprehension in FXS. We have shown previously that a Fmr1 KO rat model of FXS exhibits heightened sound sensitivity that coincided with abnormal perceptual integration of stimulus bandwidth, indicative of altered spectral processing. Frequency discrimination is a fundamental aspect of sound encoding that is important for a range of auditory processes, such as source segregation and speech comprehension, and disrupted frequency coding could thus contribute to a range of auditory issues in FXS and ASD. Here we explicitly characterized spectral processing deficits in male Fmr1 KO rats using an operant conditioning tone discrimination assay and in vivo electrophysiological recordings from the auditory cortex and inferior colliculus. We found that Fmr1 KO rats exhibited poorer frequency resolution, which corresponded with neuronal hyperactivity and broader frequency tuning in auditory cortical but not collicular neurons. Using an experimentally informed population model, we show that these cortical physiological differences can recapitulate the observed behavior discrimination deficits, with decoder performance being tightly linked to differences in cortical tuning width and signal-to-noise ratios. Together, these findings indicate that cortical hyperexcitability in Fmr1 KO rats may act to preserve signal-to-noise ratios and signal detection threshold at the expense of sound sensitivity and fine feature discrimination, highlighting a potential mechanistic locus for a range of auditory behavioral phenotypes in FXS.

The mediodorsal thalamic nucleus (MD) forms neural circuits with various brain regions, including the prefrontal cortex (PFC), the reticular thalamic nucleus (TRN), and the midbrain reticular nucleus (MRN). However, the specific roles and underlying mechanisms in auditory information processing remain unclear. Here, we perform multi-channel electrophysiological recordings in awake mice to investigate the response patterns of the MD to auditory stimuli, as well as the regulatory effects of PFC, MRN, and TRN inputs. We identify two distinct types of sound-evoked responses. The Phasic-response features a transient burst firing to the stimulus with short latency, rapidly adapting to baseline and corresponding to the onset fluctuation of the local field potential. The Sustained-response is marked by prolonged firing with longer latency and is accompanied by persistent enhancement of oscillatory power following stimulus offset. The response patterns of MD neurons remain consistent across different types of auditory stimuli. Optogenetic inactivation of the MRN suppresses both response types in the MD. The Sustained-response is attenuated by PFC inactivation but enhanced by TRN inactivation, while the Phasic-response remains unaffected by inactivation of either the PFC or TRN. Our findings expand the understanding of the MD's role in sound information integration and auditory cognitive regulation.

Electroanalytical methods are used to understand, modify, and control bionic devices. Bionic devices can record or stimulate cells to understand and/or control normal or abnormal biological processes. These devices contain electrodes that transduce electrical current within the electrical circuit into ionic current within a tissue. Despite the similarity between electroanalysis and electrophysiology, there remains a poor understanding of the relationship between the two techniques, including their methodology and theory. This paper investigates the electrochemical and acute electrophysiological recording performance of neural electrodes. A range of behaviors is achieved by modifying electrodes with the conducting polymer poly(3,4‐ethylenedioxythiophene) (PEDOT) doped with chondroitin sulfate, dextran sulfate, or para ‐toluene sulfonate. The results support previous studies showing that increased electrode area reduced total impedance below the Maxwell–Wagner relaxation frequency and thermal noise while increasing the signal‐to‐noise ratio and neural spike count. The results allowed novel investigation of relative contributions of biological and electrode properties to electrophysiological performance, with increased electrode area having a larger impact on neural population within recording range rather than reducing thermal noise. The utility of measuring electrode impedance for predicting electrophysiological performance is mainly for an indirect measure of electrode area. The results provide insight into noise sources from electrophysiological recordings and limitations in cable theory in neuroscience.

Extracellular recordings in freely moving mice, especially those with movable electrodes (microdrives), are crucial for understanding brain function. However, existing microdrives are often heavy, expensive, fragile, and unsuited for long-term studies with multichannel recordings. The OptoDrive is a new, lightweight (3.2 g), low-cost system for chronic neural recordings and optogenetic manipulation in mice. It features a detachable, 16-channel tungsten-wire electrode assembly with a 3 mm stroke (15 μm step displacement) and an integrated optical fiber. This system enables repeated implantation and explantation without surgery, requiring only gas anesthesia. The OptoDrive has demonstrated stable recordings from the lateral hypothalamus of freely behaving mice for nearly 1 month and successful optogenetic silencing of neuronal activity. In conclusion, OptoDrive offers a cost-effective, compact solution for long-term electrophysiology and optogenetics in freely moving mice.

Advancements in the field of implantable neurotechnologies have enabled the integration of hundreds of microelectrodes on ultra-thin and flexible substrates. Besides implantable components, also connectors, headstages and cables have to comply with the high-count demand, resulting in a complex and compact chain with reduced line spacing and smaller safety margins. Here, we show that epicortical recordings acquired from anesthetized rat brains with a state-of-art neural acquisition system are undoubtedly compromised by crosstalk, with signal coherence maps exhibiting a strong dependency to the routing layout. A crosstalk back-correction algorithm is developed, allowing to infer on how signals would look like under a zero-crosstalk scenario. We found that signal coherence between closely routed channels effectively drops after correction, corroborating crosstalk contamination. Our work stresses the importance of validating recorded data against the routing layout as a crucial step of data quality control, helping to come closer to ground truth data.

Distinguishing between nectar and non-nectar odors is challenging for animals due to shared compounds and varying ratios in complex mixtures. Changes in nectar production throughout the day and over the animal’s lifetime add to the complexity. The honeybee olfactory system, containing fewer than 1000 principal neurons in the early olfactory relay, the antennal lobe (AL), must learn to associate diverse volatile blends with rewards. Previous studies identified plasticity in the AL circuits, but its role in odor learning remains poorly understood. Using a biophysical computational model, tuned by in vivo electrophysiological data, and live imaging of the honeybee’s AL, we explored the neural mechanisms of plasticity in the AL. Our findings revealed that when trained with a set of rewarded and unrewarded odors, the AL inhibitory network suppresses responses to shared chemical compounds while enhancing responses to distinct compounds. This results in improved pattern separation and a more concise neural code. Our calcium imaging data support these predictions. Analysis of a graph convolutional neural network performing an odor categorization task revealed a similar mechanism for contrast enhancement. Our study provides insights into how inhibitory plasticity in the early olfactory network reshapes the coding for efficient learning of complex odors.

Neural dynamics are thought to reflect computations that relay and transform information in the brain. Previous studies have identified the neural population dynamics in many individual brain regions as a trajectory geometry, preserving a common computational motif. However, whether these populations share particular geometric patterns across brain-wide neural populations remains unclear. Here, by mapping neural dynamics widely across temporal/frontal/limbic regions in the cortical and subcortical structures of monkeys, we show that 10 neural populations, including 2,500 neurons, propagate visual item information in a stochastic manner. We found that visual inputs predominantly evoked rotational dynamics in the higher-order visual area, TE, and its downstream striatum tail, while curvy/straight dynamics appeared frequently downstream in the orbitofrontal/hippocampal network. These geometric changes were not deterministic but rather stochastic according to their respective emergence rates. Our meta-analysis results indicate that visual information propagates as a heterogeneous mixture of stochastic neural population signals in the brain.

Closed-loop neural control is a powerful tool for both the scientific exploration of neural function and for mitigating deficiencies found in open-loop deep brain stimulation (DBS). Here, we present a protocol for artificial intelligence-guided neural control in rats using deep reinforcement learning (RL) and infrared neural stimulation (INS). We describe steps for integrating RL closed-loop control into neuroscience and neuromodulation studies. We then detail procedures for using and deploying infrared INS in chronic DBS applications. For complete details on the use and execution of this protocol, please refer to Coventry et al.1 and Coventry and Bartlett.2.

To comprehend the volumetric neural connectivity of a brain organoid, it is crucial to monitor the spatiotemporal electrophysiological signals within the organoid, known as intra-organoid signals. However, previous methods risked damaging the three-dimensional (3D) cytoarchitecture of organoids, either through sectioning or inserting rigid needle-like electrodes. Also, the limited numbers of electrodes in fixed positions with non-adjustable electrode shapes were insufficient for examining the complex neural activity throughout the organoid. Herein, we present a magnetically reshapable 3D multi-electrode array (MEA) using direct printing of liquid metals for electrophysiological analysis of brain organoids. The adaptable distribution and the softness of these printed electrodes facilitate the spatiotemporal recording of intra-organoid signals. Furthermore, the unique capability to reshape these soft electrodes within the organoid using magnetic fields allows a single electrode in the MEA to record from multiple points, effectively increasing the recording site density without the need for additional electrodes.

The cerebellum is critical for coordinating movements related to eating, drinking and swallowing, all of which require proper control of the tongue. Cerebellar Purkinje cells can encode tongue movements, but it is unclear how their simple spikes and complex spikes induce changes in the shape of the tongue that contribute to goal‐directed movements. To study these relations, we recorded and stimulated Purkinje cells in the vermis and hemispheres of mice during spontaneous licking from a stationary or moving water spout. We found that Purkinje cells can encode rhythmic licking with both their simple spikes and complex spikes. Increased simple spike firing during tongue protrusion induces ipsiversive bending of the tongue. Unexpected changes in the target location trigger complex spikes that alter simple spike firing during subsequent licks, adjusting the tongue trajectory. Furthermore, we observed increased complex spike firing during behavioural state changes at both the start and the end of licking bouts. Using machine learning, we confirmed that alterations in Purkinje cell activity accompany licking, with different Purkinje cells often exerting heterogeneous encoding schemes. Our data highlight that directional movement control is paramount in cerebellar function and that modulation of the complex spikes and that of the simple spikes are complementary during acquisition and execution of sensorimotor coordination. These results bring us closer to understanding the clinical implications of cerebellar disorders during eating, drinking and swallowing. image Key points When drinking, mice make rhythmic tongue movements directed towards the water source. Cerebellar Purkinje cells can fire rhythmically in tune with the tongue movements. Purkinje cells encode changes in the position of the water source with complex spikes. Purkinje cell simple spike firing affects the direction of tongue movements. Purkinje cells that report changes in the position of the target can also adjust movements in the right direction.

The manner in which neural activity unfolds over time is thought to be central to sensory, motor and cognitive functions in the brain. Network models have long posited that the brain's computations involve time courses of activity that are shaped by the underlying network. A prediction from this view is that the activity time courses should be difficult to violate. We leveraged a brain-computer interface to challenge monkeys to violate the naturally occurring time courses of neural population activity that we observed in the motor cortex. This included challenging animals to traverse the natural time course of neural activity in a time-reversed manner. Animals were unable to violate the natural time courses of neural activity when directly challenged to do so. These results provide empirical support for the view that activity time courses observed in the brain indeed reflect the underlying network-level computational mechanisms that they are believed to implement.

Recognition of vocal identity is essential for primate social behavior, yet the neural mechanisms supporting this ability remain poorly understood. Cognitive models suggest that the brain may represent identity based on an average voice prototype. Here, we tested this hypothesis by recording spiking activity from single neurons in an fMRI-localized voice-selective region of the anterior temporal lobe in two rhesus macaques. Using synthetic “coo” calls systematically morphed along identity trajectories from “anti-voices” to caricatures, we found that early neuronal responses (∼100 ms post-stimulus) exhibited robust V-shaped tuning to distance from the average voice. This pattern reflects an energy-efficient representational strategy in which neural activity increases with identity distinctiveness, also applied to face stimuli. Unexpectedly, a distinct subpopulation of neurons showed a delayed (∼200 ms) rebound response selectively enhanced for the average voice. These findings reveal two temporally dissociable mechanisms for vocal identity encoding in primate auditory cortex, suggesting that the brain emphasizes both distinctiveness and prototypicality to support efficient and flexible voice recognition. Significance statement Recognizing who is speaking is fundamental to primate social life, yet how the brain encodes vocal identity remains poorly understood. Cognitive theories suggest that the brain may represent identity relative to an internal “average” voice, but direct neural evidence has been lacking. Here, we recorded single-neuron activity in a voice-selective brain region of rhesus macaques and found two distinct coding strategies: one that increases neural activity with identity distinctiveness, and another that selectively enhances the average voice itself. These results reveal dual mechanisms for vocal identity encoding in the primate brain, highlighting how the brain balances flexibility and efficiency in social communication.

Micro-electrocorticography (µECoG) electrodes have emerged to balance the trade-off between invasiveness and signal quality in brain recordings. However, its large-scale applicability is still hindered by a lack of comparative studies assessing the relationship between ECoG and traditional recording methods such as penetrating electrodes. This study aimed to compare somatosensory evoked potentials (SEPs) through the lenses of a µECoG and an intracortical microelectrode array (MEA). The electrodes were implanted in the pig’s primary somatosensory cortex, while SEPs were generated by applying electrical stimulation to the ulnar nerve. The SEP amplitude, signal-to-noise ratio (SNR), power spectral density (PSD), and correlation structure were analysed. Overall, SEPs resulting from MEA recordings had higher amplitudes and contained significantly more spectral power, especially at higher frequencies. However, the SNRs were similar between the interfaces. These results demonstrate the feasibility of using µECoG to decode SEPs with wide-range applications in physiology monitoring and brain–computer interfaces.

Before selecting a saccadic target, we often acquire partial information about the location of the forthcoming target and preselect a region of visual space even before the target becomes visible. To determine whether the supplementary eye field (SEF) represents information signifying the potential target space, we examined neuronal activity in the SEF of monkeys performing a behavioral task designed to isolate the process of visuospatial preselection under uncertainty from the process of selecting a specified location. Our data showed that the activity of SEF neurons represented information about the potential target space instructed by symbolic cues. Increased activity of visuospatially selective SEF neurons encoded the potential target space, which could be a mechanism facilitating subsequent selection of an appropriate target. Furthermore, electrical stimulation of the SEF during the preselection period disrupted subsequent target selection. These results demonstrate that the SEF contributes to the preselection of potential target spaces based on partial information.

Here, we characterized the dynamics of sleep spindles, focusing on their damping, which we estimated using a metric called oscillatory-Quality (o-Quality), derived by fitting an autoregressive model to electrophysiological signals, recorded from the cortex in mice. The o-Quality of sleep spindles correlates weakly with their amplitude, shows marked laminar differences and regional topography across cortical regions, reflects the level of synchrony within and between cortical networks, is strongly modulated by sleep-wake history, reflects the degree of sensory disconnection, and correlates with the strength of coupling between spindles and slow waves. As most spindle events are highly localized and not detectable with conventional low-density recording approaches, o-Quality thus emerges as a valuable metric that allows us to infer the spread and dynamics of spindle activity across the brain and directly links their spatiotemporal dynamics with local and global regulation of brain states, sleep regulation, and function.