bioRxiv Subject Collection: Neuroscience's Journal

Parallel morphological and functional development in the Xenopus retinotectal system
The retinotectal projection in Xenopus laevis is topographically organized. During the early development of the Xenopus visual system, the optic tectum increases considerably in volume, and retinotectal axons and dendrites undergo extensive activity-dependent remodeling. We have previously observed marked changes in the three-dimensional layout of the tectal retinotopic functional map over the course of a few days. This raised the question of whether such functional reorganization might be attributable to the migration and structural remodeling of tectal neurons as the brain grows. To examine changes in map topography in the context of individual tectal neuron morphology and location, we performed calcium imaging in the optic tecta of GCaMP6s-expressing tadpoles in parallel with structural imaging of tectal cells that were sparsely labelled with Alexa 594-dextran dye. We performed functional and structural imaging of the optic tectum at two developmental time points, recording the morphology of the dextran-labelled cells and quantifying the changes in their positions and the spanning volume of their dendritic fields. Comparing anatomical growth to changes in the functional retinotopic map at these early stages, we found that dendritic arbor growth kept pace with the overall growth of the optic tectum, and that individual neurons continued to receive widespread visual field input, even as the tectal retinotopic map evolved markedly over time.

Mapping Macaque to Human Cortex with Natural Scene Responses
Neuroscience has long relied on macaque studies to infer human brain function, yet identifying functionally corresponding brain regions across species and measurement modalities remains a fundamental challenge. This is especially true for higher-order cortex, where functional interpretations are constrained by narrow hypotheses and anatomical landmarks are often non-homologous. We present a data-driven approach for mapping functional correspondence across species using rich, naturalistic stimuli. By directly comparing macaque electrophysiology with human fMRI responses to 700 natural scenes, we identify fine-grained alignment based on response pattern similarity, without relying on predefined tuning concepts or hand-picked stimuli. As a test case, we examine the ventral face patch system, a well-studied but contested domain in cross-species alignment. Our approach resolves a longstanding ambiguity, yielding a correspondence consistent with full-brain anatomical warping but inconsistent with prior studies limited by narrow functional hypotheses. These findings show that natural image-evoked response patterns provide a robust foundation for cross-species functional alignment, supporting scalable comparisons as large-scale primate recordings become more widespread.

Neural sampling from cognitive maps supportsgoal-directed planning and imagination
Imagining plans for solving problems is a cornerstone of our higher cognition. We can even create plans for reaching goals that we never encountered before. However, the ways the brain addresses goal-directed imagination is largely unknown and current AI methods provide limited solutions to this problem. Here, we introduce a novel computational model -- the generative cognitive map learner (GCML) -- that successful addresses goal-directed imagination and provides a new hypothesis about the brain mechanisms supporting it. The GCML uses stochastic samples from learned cognitive maps -- data structures in which brains encode learned relations between actions and states -- to form trajectories towards both known and novel goals. In a series of simulations, we show that the GCML provides surprisingly effective heuristic solutions to spatial navigation tasks, generic problem solving tasks, and compositional tasks that require generating imagined trajectories to novel goals. The sampled trajectories capture key characteristics of hippocampal replay activity, which provide a candidate mechanism for goal-directed imagination in the brain. Since the GCML only requires simple biologically plausible local synaptic plasticity and shallow neural networks, it can be readily implemented in energy-efficient neuromorphic hardware.

Breaking Balance: Encoding local error signals in perturbations of excitation-inhibition balance
Effective learning in neural circuits requires credit assignment, linking local synaptic changes at individual neurons to distant behavioral outcomes, a challenge not addressed by classic Hebbian plasticity. While machine learning models rely on explicit local error signals to guide weight updates, we do not understand the underlying feedback mechanisms in neurobiology. Here, we propose a simple and biologically plausible solution: local deviations from precise E/I balance encode error signals that instruct synaptic plasticity. Using a computational model derived from an adaptive control theory framework, we show that targeted feedback to inhibitory interneurons perturbing E/I balance can produce neuron or assembly-specific error signals that support credit assignment in multilayer networks. Simulations demonstrate that such a balance-controlled plasticity mechanism is consistent with phenomenological local plasticity models while enabling online learning in hierarchically organized networks. Further, we show this framework is consistent with key features of dis-inhibitory microcircuit dynamics during in-vivo learning experiments. These results suggest that the brain may exploit E/I balance not just for stability, but as a substrate for error-driven learning.

Cell-type-specific synaptic scaling mechanisms differentially contribute to associative learning
Excitatory synaptic scaling regulates network dynamics by proportionally adjusting excitatory synaptic strengths after sensory perturbations. During associative learning, blocking excitatory scaling in conditioned taste aversion paradigms prolongs generalized aversive responses and delays memory specificity. Recent evidence also implicates inhibitory synaptic scaling in the regulation of network dynamics. Specifically, parvalbumin (PV)-expressing inhibitory neurons, targeting perisomatic regions of excitatory (E) pyramidal neurons, and somatostatin (SST)-expressing neurons, targeting distal dendrites, exhibit distinct scaling responses. This leaves open the question of how complex plasticity mechanisms regulate recurrent excitatory-inhibitory circuit dynamics in associative learning. Using computational approaches, we demonstrate that Hebbian plasticity drives memory generalization to novel stimuli not presented during conditioning. Following conditioning, diverse synaptic scaling mechanisms progressively induce memory specificity, which can be regulated by top-down inputs. Our results reveal that, in the absence of excitatory scaling, PV-to-E scaling can effectively compensate and rescue memory specificity, highlighting the presence of degenerate mechanisms in the brain. Notably, in the process of establishing memory specificity, excitatory scaling and PV-to-E scaling function synergistically, while concurrently opposing SST-to-E scaling. The synergistic and antagonistic plasticity mechanisms are orchestrated to shape the temporal evolution of memory representations, from generalized to precise.

Alzheimer's subtypesA supervised, unsupervised, multimodal, multilayered embedded recursive (SUMMER) AI study
Since Alzheimer's disease (AD) is a heterogeneous disease, different subtypes may have distinct biological, genetic, and clinical characteristics, requiring tailored interventions. While several proposed subtypes of AD exist, there is still no clear consensus on a definitive classification. By leveraging complementary AI approaches, including supervised and unsupervised learning, within a recursive pipeline (SUMMER) that integrates multimodal datasets encompassing MRI measurements, phenotypes, and genetic data, our goal was to generate robust scientific evidence for identifying AD subtypes. Data was downloaded from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database and included neuroimaging data (MRI), genetics (SNPs), clinical diagnosis, and demographics. 1133 European American participants' images, aged 55-95, were included in this study. The analysis was multi-fold, where the first step involved applying an unsupervised application to a subset of the MRI sample (AD + cognitively normal (CN) aged matched groups, 100 men aged 68-85 years, and 76 women aged 68-85 years). The MRI brain gray matter was segmented into 44 regions of interest (ROIs) according to a standard atlas, and 618 features were extracted, including ROI voxel intensity measurements such as minimum, maximum, and histogram variables. Results identified a cluster of subtype AD men and a cluster of subtype AD women that were distinct from the rest of their respective samples. In the next step, the integrity of the identified subtype AD clusters was investigated using the XGBoost supervised machine learning application with genetic features (SNPs, N=36,724) and labels: the identified subtype AD cluster vs. the rest of the sample, stratified by sex. A significant AD subtype men model (accuracy=0.85, F1=0.72, AUC=0.83) and a significant women AD subtype model (accuracy=0.81, F1=0.81, AUC=0.81) were built, confirming the homogeneity of the isolated AD subtype clusters. Discriminative biomarkers were extracted from the significant models, including selected ROIs and SNPs. Finally, the subtype models were tested on an unseen subset of ADNI data. The genetic-based models identified clusters of AD subtype participants consisting of 34% of the men AD group and 47% of the women AD group. Phenotypic analysis indicates that lower body weight was associated with the women's AD subtype. Complex diseases like AD demand a sophisticated, multimodal approach for precise diagnosis. Effectively identifying disease subtypes enhances the potential for personalized treatment, ultimately improving patient outcomes.

Improved Source Localization of Auditory Evoked Fields using Reciprocal BEM-FMM
We apply our recently introduced technique of reciprocal magnetoencephalographic (MEG) source estimation via the boundary element fast multipole method (reciprocal BEM-FMM) to localize auditory evoked fields (AEFs). We compare our results with the source estimates of MNE-Python against simulated N1m components of AEFs, as well as experimental data for a cohort of 7 participants subjected to binaural auditory stimulation. Previous comparisons of reciprocal BEM-FMM with MNE focused on evoked somatosensory fields, which produced results of similar quality. In this work we show that the localization of AEFs using high-resolution reciprocal BEM-FMM is significantly better in terms of accuracy and focality than those estimates of the low resolution 3-layer BEM of MNE. Our findings suggest that the use of high-resolution models plays a significant role in the quality of source estimates and that these may provide improvements in a wide range of applications.

A Large Electroencephalogram Database of Freewill Reaching and Grasping Tasks for Brain Machine Interfaces
Brain machine interfaces (BMIs) offer great potential to improve the quality of life for individuals with neurological disorders or severe motor impairments. Among various neural recording modalities, electroencephalogram (EEG) is particularly favorable for BMIs due to its noninvasive nature, portability, and high temporal resolution. Existing EEG datasets for BMIs are often limited to experimental settings that fail to address subjects' freewill in decision making. We present a large EEG dataset, containing a total of 6808 trials, recorded from 23 healthy young adults (8 females and 15 males with an age range from 18 to 24 years) while performing reaching and grasping tasks, where the target object is freely chosen at their desired pace according to their own will. This EEG dataset provides a realistic representation of reaching and grasping movement, making it useful for developing practical BMIs.

Pain and touch differentially modulate corticospinal excitability, independent of afferent inhibition
Pain can profoundly impact motor functioning to support self-preservation but is also associated with motor and somatosensory disturbances. Despite considerable research exploring the influence of pain and touch on motor and sensorimotor processes, the nature of this relationship remains elusive. Specifically, it is uncertain whether pain and touch modulate motor processes independently of each other or are interconnected. Across two experiments, an afferent inhibition (AI) paradigm was tested to probe the effects of tactile and nociceptive inputs on corticospinal processes and sensorimotor interactions. In Experiment 1 (N=20), the effect of electrocutaneous stimulation duration (0.2 vs. 0.4 ms) on transcranial magnetic stimulation (TMS)-induced corticospinal excitability (CSE) was assessed using a short and long-latency AI paradigm. A single electrocutanous stimulus was delivered to the left index finger before single pulse-TMS over the right-first dorsal interosseous (FDI) motor hotspot at one of five delays (15, 25, 35, 45, 60 or 160 ms). In Experiment 2 (N=20), the same paradigm was used to examine if this effect of sensorimotor interaction is changed when moderate tonic heat pain is delivered to the forearm. Significant AI was observed in both experiments at delays of 25, 35 and 160 ms, with afferent facilitation at 60 ms. This effect was not influenced by the duration of afferent stimulation (Experiment 1) nor by the presence of heat pain (Experiment 2). However, we found a significant reduction in CSE in painful compared to painless conditions, indicating that while tonic pain modulates CSE, tactile afferent inhibition remains unaffected. This supports the notion that pain has a direct (inhibitory) effect on motor output; however, in this context, tactile sensorimotor interactions remain unaltered.

Multivariate Bayesian Inversion for Classification and Regression
We propose the statistical modelling approach to supervised learning (i.e. predicting labels from features) as an alternative to algorithmic machine learning (ML). The approach is demonstrated by employing a multivariate general linear model (MGLM) describing the effects of labels on features, possibly accounting for covariates of no interest, in combination with prior distributions on the model parameters. ML "training" is translated into estimating the MGLM parameters via Bayesian inference and ML "testing" or application is translated into Bayesian model comparison - a reciprocal relationship we refer to as multivariate Bayesian inversion (MBI). We devise MBI algorithms for the standard cases of supervised learning, discrete classification and continuous regression, derive novel classification rules and regression predictions, and use practical examples (simulated and real data) to illustrate the benefits of the statistical modelling approach: interpretability, incorporation of prior knowledge, and probabilistic predictions. We close by discussing further advantages, disadvantages and the future potential of MBI.

Lasting Increases in Neuronal Activity and Serotonergic Receptor Expression Following Gestational Chlorpyrifos Exposure
Perinatal exposure to the organophosphorus insecticide chlorpyrifos (CPF) is associated with an increased incidence of neurodevelopmental disorders, such as autism spectrum disorder. While these behavioral detriments have been modeled in rodents, the underlying functional alterations in the developing brain are largely unknown. Previous reports using a rat model have identified alterations to both inhibitory synaptic transmission and serotonergic (5-HT) receptor binding in the cortex following developmental CPF exposure. Here, we use a rat model of gestational CPF exposure to investigate whether this altered inhibitory activity is driven by increased spontaneous firing of inhibitory interneurons and altered 5-HT receptor expression. Using cell-attached ex vivo electrophysiology in young rats of both sexes, we identified a significant increase in the number of spontaneously firing neurons in the somatosensory cortex of CPF-exposed offspring. Analysis of action potential metrics identified a sub-set of these neurons as fast-spiking parvalbumin (PV) interneurons. Immunohistochemical labeling of c-Fos, a marker of neuronal activity, further revealed a pronounced increase in activity of neurons of the somatosensory cortex in both juvenile and adult rats that had been gestationally exposed to CPF. Finally, RNAscope in situ hybridization showed an increase in the expression of the inhibitory receptor 5-HT1B in PV neurons. The data here demonstrate that gestational exposure to CPF results in persistent hyper-excitation of the somatosensory cortex, potentially through increased expression of the receptor 5-HT1B and resulting disinhibition. These neurophysiological effects may contribute to the established behavioral outcomes resulting from gestational exposure to CPF and offer guidance for novel preventative interventions.

Habenula alterations in resting state functional connectivity among autistic individuals
Background: The reward-based theoretical framework of autism suggests that altered reward circuitry contributes to core symptoms. Recent prior research has revealed autism-related structural alterations in the habenula, a small epithalamic structure associated with motivation and emotion; however, potential alterations in functional connectivity (FC) remain unexplored. Methods: Anatomical and resting state functional magnetic resonance imaging (rs-fMRI) data were accessed for 1,584 participants (N=705 autism; mean age: 16.26 plus-or-minus sign 8.15 years) in the Autism Brain Imaging Data Exchange (ABIDE). To investigate habenula alterations, we conducted a whole-brain resting state FC analysis, followed by regression analyses to explore age and brain-behavior interactions. Results: Across the entire sample, extensive habenula connectivity was observed within the midbrain dopaminergic reward system. Compared to neurotypical (NT) controls, autistic participants exhibited significantly increased habenular connectivity with the right middle temporal gyrus and bilateral superior temporal gyri. From childhood to early adulthood, habenula FC increased in autistic adolescents, and inversely decreased in NTs, with the left culmen and left parahippocampus. Between groups, habenula hyperconnectivity was inversely associated with behavioral scores for social motivation, executive functioning, and daily living skills, but not social communication. Conclusions: This study provides novel evidence of habenula connectivity alterations in autism, highlighting atypical FC with the auditory cortex. Further findings suggest that habenula circuitry develops differently among autistic adolescents, with links between habenula hyperconnectivity and motivation and adaptive behaviors. Taken together, these results contribute to emerging evidence that the dopaminergic reward system may play a critical role in the pathophysiology of autism.

Quantifying cortical maturational aspects during different vigilance states in preterm infants by advanced EEG analysis
Preterm birth is associated with numerous neurodevelopmental adverse outcomes, even in the absence of acquired lesions, as it occurs during a critical period of brain development. Clear organization of vigilance states can be recognized from 30-32 weeks postmenstrual age (PMA). In this study, we investigated whether spatial and temporal properties of neuronal oscillatory dynamics (i.e., phase synchronization, bistability, and cross-frequency coupling) during different vigilance states provide insights into cortical maturation in preterm infants born very low birth weight (VLBW) at low neurological risk and devoid of detectable brain lesions. We analyzed artifact-free video-polysomnographic data from 11 VLBW preterm infants (PMA at recording: 33.0 {+/-} 1.6 weeks) who underwent brain MRI at term-equivalent age. For each vigilance state, we computed the weighted Phase Lag Index (wPLI), Bistability Index (BiS), and Phase-Amplitude Coupling (PAC), both globally and across anterior and posterior regions, and examined their correlation with PMA at recording. wPLI, BiS, and PAC showed specific trends across vigilance states. BiS and PAC exhibited posterior-to-anterior differences and correlated with PMA. Our study suggests that these electrophysiological markers, particularly BiS and PAC, may serve as indices to monitor aspects of cortical maturation in VLBW at low neurological risk.

A Unique Role for the Hippocampus in Cue Integration During Human Spatial Navigation
A central question in cognitive neuroscience is how the brain integrates distinct sensory and perceptual inputs to improve cognitive performance. This study investigated this question in the context of human spatial navigation by combining high-field 3T fMRI and desktop virtual reality. Participants encoded and retrieved spatial locations using either landmarks alone (landmark condition), visual self-motion cues alone (i.e., optic flow; self-motion condition), or both cues together (combination condition). Participants behaviorally benefited from cue integration. fMRI analyses revealed a cue integration effect in the hippocampus, which displayed positional coding only in the combination condition, and only in participants who showed behavioral benefits from cue integration. Furthermore, hippocampal positional coding was more strongly associated with actual than self-reported locations, indicating it integrates sensory-perceptual inputs from spatial cues. Our findings suggest that fundamental function of the hippocampus is to integrate diverse sensory-perceptual inputs to support memory-guided navigation, rather than representing space per se, thereby challenging the dominant cognitive map theory in spatial navigation.

DCHS1 Modulates Forebrain Proportions in Modern Humans via a Glycosylation Change
Comparative anatomical studies of primates and extinct hominins, including Neanderthals, show that the modern human brain is characterised by a disproportionately enlarged neocortex relative to the striatum. To explore the molecular basis of this difference, we screened for missense mutations that are unique to modern humans and occur at high frequency and that alter post-translational sites. One such mutation was identified in DCHS1, a protocadherin family gene, and it was found to disrupt an N-glycosylation site in modern humans. Using CRISPR/Cas9-editing we introduced into human-induced pluripotent stem cells (hiPSCs) this ancestral DCHS1 variant present in Neanderthals and other primates, representing the ancestral state before the modern human-specific substitution. Leveraging hiPSCs-derived neural organoids, we observed an expansion of striatal progenitors at the expense of the neocortex, mirroring the anatomical distribution seen in non-human primates. We further identify the ephrin receptor EPHA4 as a binding partner of DCHS1 and show that modern human-specific alterations in DCHS1 modulate EPHA4-ephrin signalling, contributing to a gradual shift in the neocortex-to-striatum ratio - a hallmark of brain organisation in our species.

Thyroid hormone promotes fetal neurogenesis
Maternal low thyroxine (T4) serum levels during the first trimester of pregnancy correlate with cerebral cortex volume and mental development of the progeny, but why neural cells during early fetal brain development are vulnerable to maternal T4 levels remains unknown. In this study, using iPSCs obtained from a boy with a loss-of-function mutation in MCT8, a transporter previously identified as critical for thyroid hormone uptake and action in neural cells, we demonstrate that thyroid hormones induce transcriptional changes that promote the progression of human neural precursor cells along the dorsal projection trajectory. Consistent with these findings, single-cell, spatial, and bulk transcriptomics from MCT8-deficient cerebral organoids and cultures of human neural precursor cells underscore the necessity for optimal thyroid hormone levels for these cells to differentiate into neurons. The controlled intracellular activation of T4 signaling occurs through the transient expression of the enzyme type 2 deiodinase, which converts T4 into its active form, T3, alongside the coordinated expression of thyroid hormone nuclear receptors. The intracellular activation of T4 in NPCs results in transcriptional changes important for their division mode and cell cycle progression. Thus, T4 is essential for fetal neurogenesis, highlighting the importance of adequate treatment for mothers with hypothyroidism.

Selective Knockout of Murine Glutamic Acid-rich Protein 2 (GARP2) Significantly Alters Cellular Dark Noise in Rod Photoreceptors
GARP2, a glutamic-acid-rich protein found exclusively in rod photoreceptors, has been suggested to function as a structural protein, a modulator of the cGMP phosphodiesterase enzyme (PDE6), and a gating inhibitor of the rod cGMP-gated cation channel. GARP2 is a splice variant of the Cngb1 gene, which in the rods encodes the {beta}-subunit of the cyclic nucleotide-gated cation channel. Mutations in Cngb1 cause retinitis pigmentosa (RP45), and {beta}-subunit knockout mice are being studied as models of this disease. In this study, using ZFN-mediated gene editing, we have selectively eliminated GARP2 expression, while not affecting expression of the cyclic nucleotide gated cation channel {beta} -subunit, to determine its essential functions in mouse rods. The absence of GARP2 caused no consistent perturbations of retinal structure. Transiently, rod outer segment length was regionally greater than WT and infrequently misaligned, appearing parallel to the retinal pigment epithelium. Electroretinography of the knockout mice did reveal consistent functional alterations over time, seen as a reduction in the ERG response amplitudes in older mice, albeit with no significant alterations in sensitivity to light. Interestingly, single-cell patch-clamp recordings showed a significant reduction in rod photoreceptor dark noise consistent with a previously proposed role for GARP2 in binding to PDE6 and affecting its basal activity. Our results suggest a role for the GARP2-PDE6 interaction in stabilizing the PDE6 enzyme and controlling the turnover rate of cGMP in darkness, adjusting the level of dark noise and implicating an influence on the signal and noise properties of rod photoreceptors.

Somatosensory timing and cerebellar-basal ganglia beta-band interactions in Parkinson's disease
Parkinson's disease has traditionally been viewed through the lens of basal ganglia dysfunction, yet emerging research also implicates the cerebellum and its connections to the basal ganglia. To probe cerebellar and basal ganglia responses, we used a somatosensory timing paradigm, while recording magnetoencephalography. Beta-band (14-30 Hz) activity was of special interest, as we have shown that it is relevant to timing in the cerebellum. Parkinson's disease participants and controls were presented with tactile stimuli in non-jittered or jittered sequences, with the last on-time position of the sequence being omitted to probe prediction-related processes. We find prediction-related differences in the cerebellum and the basal ganglia before the omissions, but no differences in cerebellar evaluation responses after the omissions. In Parkinson's disease participants, basal ganglia activity favors jittered sequences over non-jittered, whereas the reverse is seen in the controls. Moreover, the alterations in cerebellar beta-band responses correlate with symptom severity in Parkinson's disease participants. Finally, we find an interaction in functional connectivity within a sensory-integration network between the two participant groups and the regularity of the stimuli. We interpret these findings as indicating that the cerebellum may play a compensatory role in predictive timing in Parkinson's disease and that in challenging conditions, the jittered condition, the wider sensory-integration network is harder to work in Parkinson's disease in trying to predict upcoming stimuli.

No Effect of Chronic or Acute Pain on Working Memory in the Sternberg Task
Memories of pain can last a lifetime, preventing future injuries, but this comes at the expense of remembering other concurrent experiences. For acute pain this cost is outweighed by the benefits but when pain is chronic, pain memory benefits are limited and may even contribute to maintenance of the disorder. Here we investigated two hypotheses, (1) pain takes up slots in working memory or (2) pain induces arousal above optimum levels at high task difficulty, leading to a decrease in performance. To do this, we used the Sternberg Task of working memory, in which the participant plays repeated trials where they are shown different sets of numbers and then asked to identify whether probe number was in the set or not. The Sternberg Task is ideal for testing the two hypotheses by looking at pain-related changes in accuracy and response time. There was a replication of the response time increase with set size, as well as the effect of older age on working memory and pain threshold. However, we saw no pain effect on either response time, accuracy, or the relationship between these parameters and set size with either chronic pain or an acute painful thermal stimulus. Together, this suggests that pain does not impair working memory in the Sternberg Task.

Neural Population Mechanisms for Flexible Sensorimotor Control
Modern large-scale recordings have revealed that motor cortex activity during reaching follows low-dimensional dynamics, thought to reflect sensorimotor computations. However, the origin of these patterns, and how they flexibly reorganize across different tasks remain unclear. Here we demonstrate that the key features of neural activity naturally emerge in a linear model combining a random network with a biomechanical system. Remarkably, this model shows how a fixed network can flexibly control different behaviors by optimally mapping sensory and internal state onto task-specific network inputs. Finally, analytical decomposition of the controller reveals that low-dimensional network dynamics directly follows from propagation of low-dimensional feedback signals from the biomechanical plant through the network. These results provide a biologically plausible mechanism for flexible motor control in the nervous system which directly links neural population dynamics to behavior.

Respiratory vocal coordination increases as zebra finches prepare to sing
Complex vocalizations like human speech and bird song require precise coordination between respiration and vocalizations with most vocalization being produced during the exhalation phase of respiration. However, when this coordination is established each time animals vocalize remains poorly understood. Here, we addressed this question by recording respiratory pressure in adult, male, zebra finches during singing. Zebra finches begin their song bouts by repeating a short vocalization called an introductory note (IN), before producing song and these INs have been hypothesized to reflect motor preparation for song. We found that each IN is associated with a large amplitude expiratory pulse. Birds that did not produce INs had a few silent, large amplitude, IN-like, expiratory pulses just before starting song. Expiratory pressure and inspiratory pressure increased with successive repeats of the IN (silent or vocalized). Further, for birds with INs, coordination between respiration and vocalization increased with vocalizations beginning earlier in the exhalation and ending later in the exhalation, such that more of the expiratory pulse was filled with the vocalizaton. Overall, these results show that respiratory vocal coordination improves with each repetition of the IN and suggest that INs reflect preparation that involves coordination of respiratory and vocal neural circuitry.

Human gloss perception reproduced by tiny neural networks
A key goal of vision science is to uncover the computations involved in perceiving visual properties like colour, curvature, or glossiness. Here, we used machine learning as a data-driven tool to identify potential computations of gloss perception. We generated thousands of object images using computer graphics, varying lighting, shapes, and viewpoints, and experimentally measured perceived glossiness for each image. Observers showed curious patterns of agreement and disagreement with the physical reflectance in their gloss estimates, yet their judgments remained highly consistent both within and across individuals. We then compared two sets of neural networks: one set trained to mimic the human responses (human-like networks) and another trained on physical labels to approximate physical reality (ground-truth networks). We progressively reduced the size of the networks to identify the minimum computations capable of meeting the two objective functions. While quite deep networks were required to estimate physical reflectance, we found that surprisingly shallow networks, with as few as three convolutional layers, could accurately replicate human gloss judgments. Indeed, even a miniscule network with just a single filter could predict human judgments better than the best ground-truth network. The human-like networks also successfully predicted some known perceptual gloss effects beyond the training range. Our findings suggest that humans do not judge material properties through complex computations such as inverse optics; instead, gloss perception arises from simpler computations useful for other visual tasks.

D-amphetamine alters the dynamic ECoG activity distribution patterns in the rat neocortex
Amphetamine has widespread effects on multiple neurotransmitter systems, potentially altering the physiological connectivity and network dynamics across various regions of the brain. In this study, we investigated the effects of D-amphetamine using our previously published approach where electrocorticogram (ECoG) recordings from eight cortical areas provided a coarse estimation of the global activity distribution patterns across sets of neuron populations. Changes in these activity distribution patterns were quantified with Principal Component Analysis (PCA) and k-Nearest Neighbors (kNN) classification. We found that D-amphetamine significantly altered the activity distribution patterns both for spontaneous activity and for activity recorded during ongoing tactile stimulation. It also reduced the difference between spontaneous activity and activity during ongoing tactile stimulation, which suggests that amphetamine reduced the organization in the network activity and could potentially explain hallucinations under the influence of amphetamine. Each of these changes were distributed approximately evenly across each dimension of the principal component space. This indicates that amphetamine impacts cortical network dynamics broadly and in multifaceted ways, compatible with the system-wide presence of the receptors that amphetamine interferes with. Our data indicates that relatively low doses of D-amphetamine can induce changes in brain activity distributions that are measurable potentially also by non-invasive EEG electrodes.

Interoceptive Accuracy Modulates Electroencephalogram During Music Recall Tasks
The ability to perceive and recall music varies with individuals, depending on musical experience, age, and emotional memory. Emotional memory processing occurs in the insular cortex, which is also implicated in interoception, the perception of internal bodily states, suggesting a potential link to music recall. However, the direct relationship between interoception and music recall remains largely unexplored. We hypothesize that individual differences in music recall are influenced by interoceptive accuracy. To test this, we conducted an electroencephalogram (EEG) experiment where participants listened to and recalled both familiar and unfamiliar music. The interoceptive accuracy of participants was assessed through a heartbeat counting task. We observed greater alpha power suppression during recall periods in familiar music than in unfamiliar music. Furthermore, participants with higher interoceptive accuracy exhibited stronger alpha power suppression during music recall. These findings suggest that music recall involves interoceptive attention. Considering the role of the insular cortex in both interoception and emotional memory, it may play a critical role in the neural processes underlying music recall, which should be further investigated.

Unreliable homeostatic action potential broadening in cultured dissociated neurons
Homeostatic plasticity preserves neuronal activity against perturbations. Recently, somatic action potential broadening was proposed as a key homeostatic adaptation to chronic inactivity in neocortical neurons. Since action potential shape critically controls calcium entry and neuronal function, broadening provides an attractive homeostatic feedback mechanism to regulate activity. Here, we report that chronic inactivity induced by sodium channel block does not broaden action potentials in neocortical neurons under a wide range of conditions. In contrast, action potentials were broadened in CA3 neurons of organotypic hippocampal cultures by chronic sodium channel block and in hippocampal dissociated cultures by chronic synaptic block. Mechanistically, BK-type potassium channels were proposed to underly inactivity-induced action potential broadening. However, BK channels did not affect action potential duration in our recordings. Our results indicate that action potential broadening can occur in specific neurons and conditions but is not a general mechanism of homeostatic plasticity in cultured neurons.

Pleasure in groove is associated with neuromelanin levels in the substantia nigra of younger healthy individuals: A neuromelanin-sensitive MRI study
The pleasurable urge to move in response to music is called groove. Prior research has suggested a potential link between groove and dopamine function; however, no studies to date have directly investigated the relationship between the two. Here, we aimed to assess individual dopamine function in the substantia nigra of healthy individuals using neuromelanin-sensitive magnetic resonance imaging (NM-MRI), a non-invasive method associated with dopamine function, and to investigate the relationship between the individual dopamine proxy index and sensitivity to the groove experience. In this study, 15 younger (< 48 years) and 16 older ([greater double equals]48 years) healthy individuals participated. Participants listened to ten musical excerpts and rated the groove experience based on "pleasure" and "wanting to move." To assess whether the groove experience is related to NM levels, type of musical excerpts, and sex, we analyzed with linear mixed-effects regression models. The results showed that higher NM levels and males were associated with higher pleasure ratings in the younger group. In the "urge to move" ratings, type of musical excerpts was associated with ratings in both groups, with previous research revealing higher ratings for musical excerpts with higher groove ratings (Janata et al., 2012). Taken together, these results suggest that the "pleasure" aspect of the groove experience in younger individuals was related to dopamine levels in the SN, but may not be associated with the "urge to move". Thus, pleasure and the urge to move are likely to involve distinct dopaminergic pathways and mechanisms, warranting further investigation in this regard.

Disrupted salience network dynamics during the imagery of migraine attacks
Moderate to severe head pain is a hallmark of recurring migraine attacks. However, it is challenging to study patients during spontaneous attacks and most research on the brain mechanisms of pain in migraine patients has been limited to the processing of painful stimuli between attacks. Here, we hypothesize that the experience of a migraine attack extends beyond the response to painful stimuli and is associated with specific impairments of the salience network (SN), which integrates sensory, emotional and cognitive information in relation to salient stimuli. To test this hypothesis, we analysed the SN dynamics of a group of patients with episodic migraine in three distinct conditions: at rest during a spontaneous migraine attack (ictal phase); while performing an imagery task aiming to elicit the experience of a previous attack, during the interictal phase; and at rest, during the interictal phase. For comparison, we also studied a group of healthy controls in three matching conditions, including rest as well as an imagery task of a (non-migraine) head pain experience. We collected functional magnetic resonance imaging (fMRI) data and used a dynamic functional connectivity (dFC) analysis to examine the temporal features of the SN from a total of 78 samples. Compared to healthy controls, the SN had a significantly shorter lifetime in patients during the pain imagery task, but not during a migraine attack or interictal resting state. Our results support the disruption of the SN in migraine, and indicate that pain imagery may be a useful paradigm for isolating the emotional and cognitive aspects of pain and investigating SN dynamics.

Coherent thalamic inputs organize head direction signal in the medial entorhinal cortex
Successful navigation relies on signals that remain stable despite environmental changes. This stability can arise by constraining neuronal activity to low-dimensional subspaces. During sleep, when external input is reduced, pairwise coordination persists in the spatial navigation system, as observed for head-direction (HD) cells of the anterodorsal nucleus (ADn) and grid cells of the medial entorhinal cortex (MEC). As ADn is crucial for spatial representation in the MEC, we hypothesized that coherent HD input underlies MEC organization. To test this, we performed simultaneous recordings in the ADn and MEC during wakefulness and sleep. We found that HD cell pairs maintained stable coordination across both states, and that MEC cell coordination was partly driven by common inputs from ADn HD cells. These results suggest that MEC activity is shaped, in part, by coherent thalamic HD signals, supporting stable network organization.

Biomimetic self-regulation in intrinsically motivated robots
From weaving spiders to hibernating mammals and migratory birds, nature presents numerous examples of organisms exhibiting extraordinary autonomous behaviors that ensure their self-maintenance. However, physiological needs often interact and compete. This requires living organisms to handle them as a coordinated system of internal needs rather than as isolated subsystems. We present an artificial agent equipped with a neural mass model replicating fundamental self-regulatory behaviors observed in desert lizards. Our results demonstrate that this agent not only autonomously regulates its internal temperature by navigating to areas with optimal environmental conditions, but also harmonizes this process with other internal needs, such as energy, hydration, security, and mating. This biomimetic agent outperforms a control agent lacking interoceptive awareness in terms of efficiency, fairness, and stability. Additionally, to demonstrate the flexibility of our framework, we develop a "cautious" agent that prioritizes security over other needs, achieving a Maslow-like hierarchical organization of internal needs. Together, our findings suggest that grounding robot behavior in biological principles of self-regulation provides a robust framework for designing multipurpose, intrinsically motivated agents capable of resolving trade-offs in dynamic environments.

Transcriptomic analysis reveals new reparative mechanisms of SCF and GCSF - reduced neuropathology in aged APPPS1 mice
Alzheimers Disease, AD, is a neurodegenerative disease characterized by amyloid plaque deposition, tau hyperphosphorylation, neuroinflammation, and cognitive decline. Our previous studies showed that combined treatment with stem cell factor, SCF, and granulocyte colony stimulating factor, GCSF, reduces AD pathology in APPPS1 mice. This study aimed to explore the molecular mechanism underlying SCF and GCSF therapeutic effects using transcriptomic analysis. Aged APPPS1 mice received daily subcutaneous injections of SCF and GCSF or vehicle for 12 days. RNA was extracted from brain tissue on day 13 for gene chip analysis. Age - matched wild - type, WT, mice served as controls. Data were analyzed using TAC, STRING v12 {middle dot} 0, Reactome, and ShinyGO 0 {middle dot} 77. A total of 45037 differentially expressed genes, DEGs, were detected. Twenty {middle dot} seven DEGs met a[≥] 2 - fold threshold in SCF and GCSF - treated versus vehicle - treated APPPS1 mice, 89 DEGs met this threshold in APPPS1 versus WT mice. SCF and GCSF treatment upregulated six immune - related genes, S100a8, S100a9, Ngp, Lcn2, Ltf, and Camp, associated with amyloid clearance, immune cell recruitment, and repair. Pathway analysis showed downregulation of IL - 2, IL - 4, IL - 7, and EGFR1, and upregulation of IL - 17 signaling, suggesting modulation of both innate and adaptive immunity. Notably, SCF and GCSF downregulated several oncogenes, including Cbl, Akap9, Kcnq1ot1, and Snhg11, highlighting an overlap between cancer and AD - related pathways. SCF and GCSF also promoted NADPH oxidase activation via Rho GTPases and showed > 400 - fold enrichment in metal ion sequestration, indicating potential metal chelation effects. These findings suggest that SCF and GCSF treatment modifies immune and metabolic pathways, reduces AD pathology, and highlights new therapeutic targets involving inflammation, metal homeostasis, and oncogenic signaling.

Influence of neck tissue conductivities on the phrenic nerve activation threshold during non-invasive electrical stimulation
Phrenic nerve stimulation can be used as an artificial ventilation method to reduce the adverse effects of mechanical ventilation. Detailed computational models and electromagnetic simulations are used to determine appropriate stimulation parameters. Therefore, tissue parameters have to be selected, but they vary widely in the literature. Here, we evaluated the phrenic nerve activation threshold using minimum and maximum electrical conductivity values found in the literature of each modeled neck tissue type. To calculate the phrenic nerve activation threshold, an anatomical detailed finite element model of the neck and a biophysiological nerve model were used. Considerable changes in nerve activation thresholds were found for the following tissue conductivities (with decreasing effects): muscle, skin, soft tissue, subcutaneous fat, and nerve tissue. Changes in the nerve activation threshold due to changes in skin conductivity occurred due to the bridging effect, which is an unwanted and avoidable effect during stimulation. In conclusion, fat, muscle, nerve, and soft tissue require the most accurate tissue properties and geometric representation within the model.

SARM1 is an essential component of neuronal Parthanatos
The NAD+ hydrolase SARM1 is the central executioner of pathological axon degeneration. SARM1 is allosterically activated by an increased NMN/NAD+ ratio resulting from depletion of NAD+ or accumulation of its precursor, NMN, typically due to loss of the labile NAD+ synthetase NMNAT2 following axon injury. Another NAD+ hydrolase, PARP1, is hyperactivated by DNA damage, triggering the Parthanatos cell death pathway. We demonstrate that multiple mechanistically-distinct DNA-damaging agents lead to SARM1 activation and axon degeneration following PARP1 activation. Remarkably, SARM1 is required for key steps downstream of PARP1 activation by DNA damage that are pathognomonic of Parthanatos, including mitochondrial depolarization, nuclear translocation of AIF (apoptosis-inducing factor), and cell death. Moreover, SARM1 mediates glutamate excitotoxicity, a clinically significant pathomechanism attributed to Parthanatos. The identification of SARM1 as an essential component of neuronal Parthanatos, a major contributor to cell death in neurodegenerative disease, greatly expands the potential clinical utility of SARM1 inhibitors.

Anti-amyloid beta therapy resolves stroke recovery impairment caused by Alzheimer's disease
Stroke and dementia are common comorbidities and a growing concern causing disability in aging societies worldwide. Although anti-amyloid beta (anti-A{beta}) antibodies have recently been anticipated to relieve preclinical Alzheimer's disease pathology, we discovered that post-stroke administration of anti-A{beta} antibodies restored neural repair for stroke recovery impeded by cerebral A{beta} accumulation. Neuronal recovery-associated gene expression for stroke recovery was considerably impaired even by slight A{beta} accumulation in murine and human brain. Slight A{beta} accumulation had less impact on neurons without stroke but caused a unique myeloid immunity after an ischemic stroke that enhanced the inflammatory cascades impeding neural repair for stroke recovery. Aducanumab administration after ischemic stroke prevented formation of this malignant myeloid immunity, resolving the impairment of post-stroke neural repair caused by cerebral A{beta} accumulation. Thus, our study has revealed the ability of anti-A{beta} therapies to restore functional recovery after a stroke with cerebral A{beta} accumulation.

Cryo-EM evidence for a common factor in Alzheimer's and other neurodegenerations
In the last seven years, cryo-EM maps of neuropathological fibrils from Alzheimer's disease and other neurodegenerations have been released by various authors. The first publication noted an unknown component coordinating with lysine residues in the protein, a finding recapitulated in many succeeding studies. Previous authors have emphasized difficulties in analysing this component, but current findings, using powerful visualisation software UCSF ChimeraX on all publicly available maps, indicate that the issue is tractable. Lysine-coordinating extra densities have common features, including a Y-shaped substructure, suggestive of a molecular factor in common, in neuropathological fibrils from a wide range of neurodegenerations and involving misfolded proteins beta-amyloid, alpha-synuclein, prion protein, tau and transmembrane protein 106B. A similar component, albeit in non-lysine environments, was found in neuropathological fibrils involving TAR DNA-binding protein 43 and TATA-binding protein-associated factor 15. The results suggest the existence of a common molecular factor, a predominantly anionic polymer, linking these diseases and raising the possibility of a unitary basis for Alzheimer's and other neurodegenerations. Based on evidence here, RNA is a feasible candidate for this putative common factor. Such findings raise the possibility of new diagnostic tests and treatments for these devastating diseases in the future.

An 'Aha!' moment precedes the strategic response to a visuomotor rotation
Strategic behaviour in sensorimotor adaptation tasks is typically modelled either as an error minimisation process or as a process of learning through trial-and-error. The former predicts a gradual reduction in error, until some asymptote, while the latter predicts behavioural exploration to discover an efficacious solution. An alternative explanation is that a sufficiently rich understanding of the task culminates in an 'Aha!' moment, which then allows the generation of a new solution. This predicts some period of perseveration in baseline behaviour, followed by an abrupt single-trial shift to a new strategic solution. To avoid obfuscation caused by the motor system, we investigate these hypotheses in a strategy-only aiming game, where participants aim and fire a cannon at targets. Most participants exhibited a period of baseline perseveration followed by a single-trial shift to good performance. We then applied the same analyses to reaching data from a visuomotor rotation task that inhibited implicit adaptation through delaying feedback presentation (Brudner et al., 2016). Similarly, we found that participants typically perseverated in reaching towards the target before suddenly switching, in a single trial, to good performance. These findings suggest that traditional descriptions of strategic behaviour are insufficient and must be updated if we want to understand how humans respond to sensorimotor perturbations.

Parkinson's-Linked Synaptojanin 1 Loss in Dopamine Neurons Triggers Synaptic Degeneration and Striatal TH Interneuron Compensation
Synaptic dysfunction is increasingly recognized as an early hallmark of Parkinson's disease (PD). Synaptojanin 1 (SJ1), a synaptic phosphoinositide phosphatase essential for synaptic vesicle recycling, is genetically linked to early-onset Parkinsonism (EOP). While germline SJ1 knockout mice are perinatal lethal, SJ1-R258Q knock-in mice recapitulate EOP-like symptoms and exhibit selective dystrophy of nigrostriatal dopamine (DA) terminals. However, these whole-body mutants limit understanding of SJ1's role in the DA system. Here, we generated DA neuron-specific SJ1 conditional KO (SJ1-DA cKO) mice. Complete SJ1 loss caused severe synaptic dystrophy throughout the striatum, indicating a cell-autonomous role of SJ1 in both substantia nigra and ventral tegmental area DA subtypes. Surprisingly, despite profound synaptic degeneration and DA deficiency, SJ1-DA cKO mice showed no overt motor deficits. Instead, we observed a robust induction of striatal tyrosine hydroxylase-positive interneurons (iTHINs), which expressed multiple DA markers and formed new connections with degenerating DA terminals, suggesting a potential local compensation. In contrast, acute SJ1 deletion in adult/aged DA neurons caused similar terminal pathology but limited iTHINs induction. Together, our findings reveal a critical role for SJ1 in maintaining DA synaptic function and uncover an adaptive striatal response to DA loss, offering insights into compensatory mechanisms relevant to PD pathogenesis.

Economic and Social Modulations of Innate Decision-Making in Mice Exposed to Visual Threats
When confronted by a predator, most animals make innate decisions with rapid reaction times - a trait shaped by natural selection to maximize survival. However, in complex and dynamic environments, fast reactions are meaningful only when grounded in accurate judgments and correct choices, which often require cognitive control. Here, we investigate how threat intensity, reward value, and social hierarchy influence behavioral decisions in foraging mice exposed to overhead visual threats. We find that threat intensity plays a dominant role in decision-making: elevated threat levels trigger robust defensive responses, including shorter latencies to flee, increased fleeing speeds, and longer escape distances. The influence of reward is context-dependent: at low threat levels, higher reward values suppress defensive responses, leading to prolonged time in the reward zone, slower speeds, and shorter escape distances. In contrast, under high-threat conditions, increasing reward value enhances sensitivity to threats, as evidenced by shorter latencies to flee and heightened vigilance. Social hierarchy further shapes decision-making, with dominant mice exhibiting greater vigilance and a stronger preference for risk-averse behaviors compared to subordinates. To quantify the decision-making process, we developed a drift-diffusion leaky integrator model that successfully captures how mice integrate threat intensity, reward value, and vigilance into their behavioral decisions. Our findings reveal the economic and social modulation of survival decisions, offering insights into the computational mechanisms underlying the interplay between instinctive reaction and cognitive control in innate decision-making.

Humans underestimate their body mass in microgravity: evidence from reaching movements during spaceflight
Astronauts consistently exhibit slower movements in microgravity, even during tasks requiring rapid responses. The sensorimotor mechanisms underlying this general slowing remain debated. Two competing hypotheses have been proposed: either the sensorimotor system adopts a conservative control strategy for safety and postural stability, or the system underestimates body mass due to reduced inputs from proprioceptive receptors. To resolve the debate, we studied twelve taikonauts aboard the China Space Station performing a classical hand-reaching task. Compared to their pre-flight performance and to an age-matched control group, participants showed increased movement durations and altered kinematic profiles in microgravity. Model-based analyses of motor control parameters revealed that these changes stemmed from reduced initial force generation in the feedforward control phase followed by compensatory feedback-based corrections. These findings support the body mass underestimation hypothesis while refuting the strategic slowing hypothesis. Importantly, sensory estimate of bodily property in microgravity is biased but evaded from sensorimotor adaptation, calling for an extension of existing theories of motor learning.

High-intensity interval exercise affects explicit sequential motor consolidation with both physical and mental practice
High-intensity interval exercise (HIIE) is known to enhance motor consolidation following physical practice (PP), but its effects on sequential motor learning (SML) through PP or motor imagery (MI) remain unclear. We examined whether HIIE modulates SML consolidation in 48 participants who learned an explicit SML task through PP or MI. Performance was assessed before and after acquisition, after HIIE or rest, and at 24 hours and 7 days. Both PP and MI improved performance, with greater gains for PP (p = 0.042), and both induced intracortical disinhibition (p = 0.03). HIIE increased BDNF (p = 0.044) and lactate levels (p < 0.001), markers typically linked to neuroplasticity, yet unexpectedly impaired SML at early (p < 0.01) and late consolidation (p < 0.05), without affecting excitability. These findings challenge the presumed coupling between exercise-induced biomarkers and behavioral gains, suggesting that HIIE may hinder consolidation when explicit components of motor learning are involved.

Absence of Systematic Effects of Internalizing Psychopathology on Learning Under Uncertainty
Difficulties in adapting learning to meet the challenges of uncertain and changing environments are widely thought to play a central role in internalizing psychopathology, including anxiety and depression. This view stems from findings linking trait anxiety and transdiagnostic internalizing symptoms to learning impairments in laboratory tasks often used as proxies for real-world behavioral flexibility. These tasks typically require learners to adjust learning rates dynamically in response to uncertainty, for instance, increasing learning from prediction errors in volatile environments. However, prior studies have produced inconsistent and sometimes contradictory findings regarding the nature and extent of learning impairments in populations with internalizing disorders. To address this, we conducted eight experiments (N = 820) using predictive inference and reversal learning tasks, and applied a bi-factor analysis to capture internalizing symptom variance shared across and differentiated between anxiety and depression. While we observed robust evidence for adaptive learning-rate modulation across participants, we found no convincing evidence of a systematic relationship between internalizing symptoms and either learning rates or task performance. These findings challenge prominent claims that learning difficulties are a hallmark feature of internalizing psychopathology and suggest that the relationship between these traits and adaptive behavior under uncertainty may be more subtle than previously thought.

Automated behavior classification of julius seizure mutants in Drosophila reveals stereotyped seizure stages with genotype specificity
Bang-sensitive (BS) Drosophila mutants exhibit a stereotyped pattern of seizure behavior after mechanical disturbances. We previously identified mutations in the julius seizure (jus) gene, formerly CG14509, can induce BS seizures. However, the behavioral manifestations of the seizure phenotype of the various jus mutants have not been fully characterized. Here, we developed a machine learning pipeline featuring LASC (Long short-term memory and Attention mechanism for Sequence Classification) for automatic phenotyping of jus mutant videos. LASC achieves 90% classification accuracy in distinguishing five phases: paralysis (P), tonic seizure (T), spasm (S), clonic seizure (C), and recovery (R). Applying the trained LASC model to multiple jus lines showed they use a common repertoire of seizure stages and followed the general P[->]T[->]S[->]C[->]R progression, but each genotype exhibited unique patterns of stage duration and transition probabilities. Remarkably, stage usage patterns are distinct among the mutant genotypes. These findings establish that while all jus mutants adhere to stereotyped behavioral rules, each allele generates a distinct signature in stage usage. This work demonstrates how advanced behavioral quantification can reveal previously hidden relationships between gene mutation and complex motor outputs. More broadly, the complete pipeline presented here can pave the way for high-throughput, automated drug screening for epilepsy.