Frontiers in Neuroscience

Musical rhythm is often experienced with a periodic beat, serving as a temporal reference for coordination with the rhythm. Thus far, models of beat processing have mainly relied on representing sensory inputs as patterns of onset timing, with limited consideration of other sensory features. Here, we challenge this view by showing that the internal representation of beat is affected by other temporal features of the stimulus beyond onset timing alone. We recorded electroencephalography (EEG) while participants listened to rhythmic sequences designed to elicit a beat. Across conditions, we manipulated the duration of the tones conveying the rhythms, while keeping all other parameters identical, including overall intensity, speed, and rhythmic pattern structure. Crucially, the beat periodicity was enhanced in neural activity with increased sound duration, even though the beat periodicity was not prominent in the acoustic features, thus ruling out basic sensory confounds. These results demonstrate the preferential role of longer sound durations in fostering temporal scaffolding processes that integrate fast rhythmic inputs into behavior-relevant internal structures such as the beat. More generally, our findings are compatible with a holistic processing account whereby a range of features beyond onset timing may be integrated into a neural representation of rhythm. Graphical Abstract: Fig. 2EEG was recorded while listeners heard rhythmic sequences eliciting a beat. Sound duration (sonic duty cycle) was varied across four conditions while speed, pattern, and intensity stayed constant. Beat-related EEG responses increased with longer sounds, and were enhanced in all conditions compared to auditory nerve model envelopes, which did not show prominent energy at the beat periodicity, ruling out sensory confounds. Results support holistic rhythm processing beyond onset timing alone. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=101 SRC="FIGDIR/small/721298v1_fig2.gif" ALT="Figure 2"> View larger version (27K): org.highwire.dtl.DTLVardef@10a0599org.highwire.dtl.DTLVardef@f5a95forg.highwire.dtl.DTLVardef@42d1ceorg.highwire.dtl.DTLVardef@dc58a7_HPS_FORMAT_FIGEXP M_FIG O_FLOATNOFigure 2.C_FLOATNO EEG and auditory nerve model output analysis based on magnitude spectrum and autocorrelation. Each row represents a duty cycle condition. The two columns on the left represent the magnitude spectrum-based analysis. The first column represents the group-level averaged magnitude spectra at a pool of fronto-central electrodes, across conditions. Beat-related frequencies are shown in red, and beat-unrelated frequencies are shown in blue. Scalp topographies of the neural activity measured at the average magnitudes of beat-related (in red circle) and unrelated (in blue circle) frequencies are represented as insets. The second column represents the normalized magnitude spectra obtained from the auditory nerve model output for each duty cycle sequence. The two columns on the right represent the autocorrelation-based analysis (for visualization purposes, only a subset of lags from 0 to 2.4 s corresponding to the pattern duration is shown). The first column represents the group-level averaged autocorrelation function measured from the same pool of fronto-central electrodes, across conditions. Beat-related lags are shown in red, and beat-unrelated lags are shown in blue. The second column represents the autocorrelation function of the auditory nerve model output for each duty cycle sequence. C_FIG

Understanding speech in noisy situations is challenging and often fails due to attentional rather than sensory deficits. We report here that neurofeedback can enhance a neural signature of selective attention to speech in a cocktail-party setting. Neural responses to speech were quantified in participants electroencephalogram (EEG) while they attended to one of two audiobooks, presented simultaneously. After each 22s-segment of speech, the participants N1 component was extracted from the temporal response function (TRF). The N1 component is known to be sensitive to attention. During a [~]49-min training session, N1 amplitude values were displayed visually to participants so that they could learn to strengthen their neural responses to target speech and minimise their responses to distracting speech. During neurofeedback training, we found an enhanced N1 component in the response to target audiobooks that was specific to EEG channels used to provide feedback, and not present in a control group that received sham feedback. At right-lateralised fronto-central channels, enhanced N1 components correlated with improvements in a measure of speech comprehension (multiple-choice content questions). These results indicate that neural responses to speech can be regulated through neurofeedback and open up new possibilities to train attentional listening in populations struggling to understand speech in noise.

Chronic tinnitus is a common condition with few effective treatments and no cure. Though inconsistent results across MRI studies of tinnitus have slowed mechanistic insight, converging evidence across animal and human studies clearly implicate auditory-system dysfunction. This paper presents a systematic, retrospective assessment of auditory-network function in chronic tinnitus across multiple fMRI datasets. Auditory network nodes were newly defined in this effort, including novel nodes in cerebellum previously linked with somatotopic representations of articulators (lobules VI, VIIIa). Auditory-network connectivity in cerebellum and superior olivary complex was reduced in chronic tinnitus, perhaps explaining the recent success of trigeminal stimulation in improving tinnitus. Auditory-network strength was also reduced, corroborating some recent studies and perhaps reflecting increased spontaneous neuronal activity reported in animal models. Together, these results suggest auditory-network dysconnectivity as a tinnitus biomarker, and that efferent cochlear pathways related to head-centric interoception may play a mechanistic role.

Organotypic slice cultures (OSCs) are widely used to study cellular properties in a functional and developmental tissue context. With the recent advent of transgenic mouse lines and viral tools we postulated that OSCs may enable the study of multicellular glial and neuroglial interactions in development, as well homeostatic and pathological conditions. Here, we made mouse cortical OSCs and used markers for oligodendroglial, microglial states and neuronal types between 1 to 28 days in vitro (DIV). The OSC was characterized by in-vivo like cortical layering, including layer 5 pyramidal neurons and produced highly robust synchronized period bursts resembling Up- and Down states. Glial cells showed a strong cortical layer- and time-dependent development pattern: in the first week (DIV 1-7), slicing-related debris clearance and developmentally restricted sparse oligodendroglial myelination created an environment with highly phagocytic, non-homeostatic microglia (assessed with CD68 and purinergic receptor P2Y12, respectively). Between DIV 14 and 21, however, slices showed stereotypical cortical myelin patterns and the emergence of a homeostatic microglia phenotype while exhibiting continued phagocytosis. Furthermore, live two-photon imaging and morphometric analyses revealed highly ramified microglia and myelinated axons with compact myelination, exceeding lamellae count compared to age-matched in vivo axons. Lastly, from DIV 28 and onwards, myelin integrity became impaired and associated with phagocytic microglia. Together, the results indicate that between DIV14 and 21 cortical OSCs are well suited for live imaging of homeostatic and activity-dependent neuron-glia interactions, bridging the gap between in vivo investigations and primary cell cultures.

The most prominent signature associated with motor execution and motor imagery is the event-related desynchronisation and synchronisation (ERD/S) in the mu and beta bands (8-30 Hz). In the context of brain-computer interfaces (BCI), this ERD/S signature is helpful for binary decisions, such as left vs. right imagery, but it is not a robust biomarker for continuous prediction, such as precisely decoding different levels of force application. This is essential for developing better BCI applications with precise dynamic force outputs. Recent studies have revealed that sensorimotor beta bursts have a stronger relationship with motor control, even at a single-trial level, than beta band power. We, therefore, investigated whether the transient nature of beta bursts provide an alternative, but robust biomarker for BCI force decoding. Here, we designed an experiment where human participants (N = 16) performed both motor execution (ME) at four force levels (10%, 25%, 50%, and 75% of maximum voluntary contraction) and imagined exerting the same, i.e. a motor imagery (MI) task, as their electroencephalogram was recorded. We observed a clear and classical ERD pattern in the motor cortex during the ME task, whereas it was less pronounced during the MI task. After extracting sensorimotor beta bursts, we observed differences in spectral burst features between motor execution and imagery including burst amplitude, spectral width, and temporal width. Moreover, different force levels were correlated with changes in the burst amplitude and burst spectral width, specifically during motor execution. Interestingly, we found that different beta burst waveforms are associated with the different force levels and conditions. This suggests that the bursts-level features could be driven by changes in the underlying beta burst waveforms. Overall, our study shows that sensorimotor beta burst can be an important piece of the puzzle to implementing precise force control in brain-computer interface-based prosthetics.

Purpose Considering that myopia is associated with thinning of the ocular coats, this study investigated the inter-relationship of retinal, choroidal and scleral thickness in foveal regions in Indian high myopes. Methods A total of 23 high myopes (spherical equivalent refraction [&le;]-6.00D) aged 16 to 35 years underwent posterior segment imaging with swept-source optical coherence tomography. The retinal, choroidal and scleral thickness was determined using semi-automated custom-designed software at sub-foveal regions. Axial length was determined using Lenstar LS 900 non-contact biometer. Results The mean plus-or-minus sign SD axial length was 30.17 plus-or-minus sign 2.23 mm, sub-foveal retinal thickness was 245 plus-or-minus sign 28 lower case Greek mum, sub-foveal choroidal thickness was 82 plus-or-minus sign 46 lower case Greek mum, and sub-foveal scleral thickness was 254 plus-or-minus sign 68 lower case Greek mum. The choroid was significantly thinner compared to the retina and sclera (p<0.001). With a 1 mm increase in axial length, there was no significant variation in sub-foveal retinal (increased by 0.86 lower case Greek mum) and scleral thickness (decreased by 4.31 lower case Greek mum, p[&ge;]0.05), but sub-foveal choroidal thickness decreased by 10.35 lower case Greek mum (p=0.02). For a 1D decrease in spherical equivalent refraction, the choroidal thickness reduced significantly (decreased by 5.88 lower case Greek mum, p<0.001), while there was no significant variation in retinal (decreased by 0.68 lower case Greek mum, p=0.55) and scleral thickness (increased by 0.13 mum, p=0.98). The association of the sub-foveal retinal, choroidal, and scleral thickness was weak and was not significant in high myopes (p[&ge;]0.10). Conclusions With increasing axial length and severity of myopia in high myopes, compared to scleral and retinal thickness, the choroidal thickness alone decreased significantly. Our findings indicate that the changes in the choroid do not necessarily reflect the changes in retinal and scleral thickness and highlight the importance of the choroid as a marker for axial elongation even in high myopes.

Brain stimulation is increasingly recognized as an effective and important therapeutic intervention for many brain diseases. Distance between the scalp and other brain regions is a pivotal variable for neurostimulation planning and the development of new techniques, but alterations in the distance between the scalp and other regions in brain diseases are largely unknown. In this study, we developed an automatic pipeline to calculate scalp-to-region distance (SRD) values from T1 MR images and applied it to a total of 1382 participants, including patients with autism spectrum disorder (ASD), Parkinsons disease (PD), Alzheimers disease (AD), and cognitively normal controls (CNs). Cloud points were uniformly sampled on the automatically extracted scalp surface and cortex surface, on which the point-wise distance maps were generated. The brain was then coregistered with the BCI-DNI atlas, and SRD value for each brain region was extracted. Analysis of covariance (ANCOVA) was performed for SRD in each brain region, with age and sex as covariates. Compared with CNs, ASD patients showed widespread SRD decreases across the brain with prominent involvement of the frontal lobe, especially the orbitofrontal cortex and adjacent regions. In contrast, in AD patients, significantly increased SRD values were observed in various regions of the frontal gyrus. No significant SRD alteration was found in PD patients after correction. The automatic SRD calculation pipeline and the different patterns of SRD alterations in these diseases might be helpful for future neurostimulation planning in clinical practice. HighlightsO_LIAutomatic pipeline enables scalp-to-region distance (SRD) measurement, facilitates brain stimulation planning. C_LIO_LIASD patients show widespread SRD decreases, especially in the orbitofrontal cortex and adjacent regions. C_LIO_LIAD patients present increased SRD in the frontal gyrus and decreased SRD in the parahippocampal gyrus. C_LI

Computational models of brain structures impacted by Parkinsons disease are useful for exploring potential therapies. We use the Hindmarsh-Rose neuronal model to simulate synchronized activity in the subthalamic nucleus, capturing key features of the pathological rhythms observed in Parkinsons disease using a relatively small network of 100 neurons. Our model incorporates unidirectional excitatory chemical synapses whose strengths evolve according to a spike-timing-dependent plasticity (STDP) rule. To account for inputs from unmodelled neurons, both uniformly distributed white noise and Poisson noise were explored. White noise produced a single stable state of synchronized neuronal activity whereas Poisson noise resulted in two stable states, one synchronized and one desynchronized. We applied coordinated reset stimulation with a rapidly varying sequence (RVS CR) to examine its ability to reduce neuronal synchrony. The neuronal population was divided into subpopulations representing distinct physical sites of stimulation, as in deep brain stimulation therapy, and phase-shifted stimuli were delivered to each subpopulation in a random sequence. We explored how stimulation frequency and the number of stimulation sites affect the efficacy of RVS CR at desynchronizing the network. We demonstrate that RVS CR efficacy is sensitive to the depression-to-potentiation ratio in the STDP rule, which may be an important parameter to tune when reconciling simulations with experimental data. Numerical simulation of neuronal networks is constrained by computational resources when models demand large networks. This work proposes a model that demonstrates similar utility with a relatively small network, enabling researchers to study pathological neuronal activity and treatments more efficiently.

Studies in rodents and humans using invasive electrophysiology have established that neural replay is a ubiquitous phenomenon in the brain that is associated with a wide range of cognitive functions, including memory, planning and decision making. Yet, invasively recording in humans remains difficult, and hence knowledge about replay in humans remains scarce. Hence, to comprehensively understand replay in humans, we need reliable approaches that can detect it non-invasively. Several main non-invasive approaches have been proposed, but we lack a full comparative validation against known ground truth signals. In this study, we present FASTIMAGES, a benchmark dataset from seventy participants with parallel fMRI (n = 40, previously published) and MEG (n=30) recordings containing known neural sequences evoked by fast visual stimulation as well as functional localizer trials. The neural sequences were elicited by five different visual stimuli shown in sequences at speeds of 132, 164, 228 and 612 milliseconds onset-to-onset intervals. Using this dataset, we investigate two existing statistical methods for sequence detection, namely Temporally Delayed Linear Modelling (TDLM, developed for MEG by Liu et al., 2021) and Slope Order Dynamic Analysis (SODA, developed for fMRI by Wittkuhn & Schuck, 2021). We examine the underlying assumptions of each method, analyse their resulting strengths and weaknesses in application to MEG and fMRI. We demonstrate that both approaches excel in their native modality (TDLM for MEG and SODA for fMRI), with comparable effect sizes given idealized conditions in this benchmark. Cross-modality transfer remains challenging. Finally, the FASTIMAGES dataset provides data with known and clearly expressed sequences and can be used to benchmark and validate future sequence detection methods under idealized conditions.

Importance: Leukodystrophies are a heterogeneous group of genetic disorders affecting the white matter of the brain, often presenting with overlapping clinical features but differing in neuroanatomical involvement. There is a critical need for quantitative tools to characterize disease burden and support diagnosis, severity stratification, and clinical trial readiness. Objective: To characterize shared and distinct neuroanatomical patterns across six genetically confirmed leukodystrophies using anatomical MRI-derived phenotypes benchmarked against brain growth charts, and to assess the utility of this methodological approach for identifying imaging biomarkers of disease severity. Design, Setting, and Participants: Cross-sectional neuroimaging study using retrospective clinical MRI data. Setting: Multicenter study incorporating data from the Global Leukodystrophy Initiative Clinical Trials Network (GLIA-CTN) and control data from the Childrens Hospital of Philadelphia. Participants: The study included 434 MRI scan sessions from 274 patients with genetically confirmed leukodystrophies (Pelizaeus-Merzbacher disease, Metachromatic leukodystrophy, Alexander disease, Aicardi-Goutieres syndrome, TUBB4A-related leukodystrophies, and POLR3-related leukodystrophy). Control MRI data (7628 scans from 7205 subjects) were drawn from the Scans with Limited Imaging Pathology cohort at the Children's Hospital of Philadelphia. Exposures: All MRI scans underwent automated segmentation using deep learning segmentation tools to derive global and regional brain volumes. Normative models of brain development ("brain growth charts") were generated for the control cohort using generalized additive models for location, scale, and shape. Centile scores were then calculated for leukodystrophy subjects to quantify deviations from typical development. Main Outcomes and Measures: Centile scores for global and regional brain volumes were compared across leukodystrophy subtypes to identify disease-specific neuroanatomical patterns and to evaluate their potential utility for severity stratification. Results: Distinct patterns of neuroanatomical deviation were observed across leukodystrophy subtypes. Certain leukodystrophies showed preferential involvement of specific cortical or subcortical regions, while others displayed more diffuse volume loss. Centile scores demonstrated potential for differentiating disease subtypes and stratifying individuals by severity. Preliminary longitudinal data suggest centile scores may also track progression over time. Conclusions and Relevance:This study demonstrates the feasibility and utility of MRI profiling of individuals with leukodystrophy using anatomical MRI-derived phenotypes benchmarked against brain growth charts. The approach enables data-driven, quantitative characterization of structural brain abnormalities, offering a scalable method for phenotyping, diagnosis, and future use in clinical trials.

IntroductionStimulation-evoked potentials (SEPs), recorded both during and after deep brain stimulation (DBS) surgery, have shown promise for guiding DBS targeting and programming. However, filtering protocols applied to stimulation trains produce an artifact we call a filter-induced oscillation (FIO) which closely mimics physiological SEPs. Hence, we outline the mechanistic origins of this distortion and describe a means of differentiating it from valid SEP activity. MethodsWe recorded in 18 patients undergoing DBS surgery targeting the subthalamic nucleus or globus pallidus internus. We stimulated target nuclei with cathode-first (CF) and anode-first (AF) pulses to record native SEPs, and in white matter tracts (null condition). Recordings were subsequently filtered to illustrate FIO. Next, we filtered harmonic frequencies of an artificial stimulation train to demonstrate FIO origins. Finally, FIO was deliberately generated in white matter recordings with a notch filter, and its behaviour contrasted with SEPs during AF and CF stimulation. ResultsFiltering stimulation trains produced FIOs that depended on filter order and corner frequency. We also showed that FIO emerges from filter-induced attenuations of harmonic frequencies which compose stimulation trains, producing oscillations of like frequency around pulses. Finally, FIOs reverse in polarity depending on AF or CF stimulation, whereas SEPs do not. ConclusionsGiven the potential for widespread adoption of SEPs in DBS targeting and programming, safe analytical protocols are imperative to avoid the induction of processing-related artifacts which can be misinterpreted as biological signals. Here we establish the necessary theory for identifying FIOs and tuning analytical pipelines to avoid their generation.

Sleep spindles are rhythmic electroencephalographic signatures of non-rapid-eye-movement sleep. Their dysregulation has been implicated in several neuropsychiatric illnesses. Spindles have a characteristic waxing and waning shape, but the cellular and circuit mechanisms controlling their shape are not well understood. Recent but sparse research has implied that sleep spindle shape becomes abnormal in post-traumatic stress disorder (PTSD). PTSD patients have dysfunctional GABAA receptors in midline thalamic regions, areas involved in the orchestration of sleep spindles. We modelled this GABAA dysfunction within thalamocortical (TC) neurons using localized CRISPR-Cas9 technology to test the hypothesis that GABA dysfunction would dysregulate sleep spindle shape and cause symptoms of PTSD, in mouse model behavioral evaluations. We found sleep spindles were shorter and abnormally shaped, having lost their characteristic waxing and waning shape, in mice with GABAA receptor knock-down in TC neurons (TC-1KD). TC-1KD mice failed to recover from learned fearful reactions following an aversive stimulus. We tested this with a contextual fear conditioning paradigm using electric foot shocks. A control group with intact GABAA receptors successfully habituated to the fear conditioned location in subsequent visits to that context without foot shocks. In contrast, TC-1KD mice never habituated, suggesting abnormally extended fearful memories. The number of inhibitory post synaptic currents in TC neurons were significantly decreased in vitro, confirming an effective knock-down. Our results imply that abnormally shaped sleep spindles may serve as a biomarker of GABAA receptor dysfunction in TC neurons which may be involved in abnormal fear processing in PTSD. We postulate GABAA receptor dysfunction in TC neurons may be underlying pathophysiology of PTSD and our findings here may inspire the development of screens, diagnostics and objective characteristics of stress related disorders, including PTSD.

Zebrafish (Danio rerio) are an important vertebrate model for vision and neuroscience research. In the larval stages, the aquatic species begins to elicit the optomotor response (OMR) to stabilize themselves in water -- a behaviour that may be exploited in the laboratory to measure visual acuity. However, up to now, the measurement of the OMR in juvenile and adult zebrafish has been limited due to their behavioural complexity. Here, we optimize a protocol to assay zebrafish aged between 4 and 9 weeks-post-fertilization, by displaying sinusoidal gratings parallel to the zebrafish eye to elicit a robust OMR. We assessed the visual spatial-frequency tuning function of an environmentally induced myopia model to confirm the sensitivity and robustness of the protocol. Additionally, we show the OMR is sensitive to the contrast and temporal resolution of the sinusoidal gratings. Furthermore, we found that the time between stimulus presentations impact the spatial-frequency tuning function likely as time is required for zebrafish to return to baseline swimming after eliciting the OMR. Finally, we found that the OMR after ten versus twenty seconds of stimulus onset appears comparable; indicating that robust OMR responses in zebrafish can be elicited through relatively short stimulus presentations. Through the experiments conducted, we present an optimized protocol specific to zebrafish. The protocol may be used to follow the progression or treatment efficacy of progressive neurological disorders including specific visual disorders and higher brain functions with visual endophenotypes. Ultimately, this protocol allows for high-throughput robust measures of visual and neural function in zebrafish.

BackgroundParkinsons disease (PD) is the second most common neurodegenerative disorder. PD currently lacks effective disease-modifying treatments, likely due to its diverse clinical features and underlying neuropathology. The vascular role in PD is emerging, with vascular mechanisms increasingly implicated, yet the literature remains conflicted, motivating large-data analyses with greater statistical power. White matter lesions (WML) are an accepted imaging marker of small vessel disease. Accurate automated WML segmentation techniques are crucial for large-scale studies in PD due to the impracticality of manual segmentation for extensive datasets and to ensure consistency. Evaluation of the optimum approach in PD for large-scale analysis is lacking. This study aimed to evaluate various automated WML segmentation algorithms to determine the most accurate and reliable method, among those selected, for assessing WML for multi-site large data analysis in PD. MethodsWe assessed whole-brain volumetric T1-weighted and FLAIR images from 201 PD patients (mean age, 66.6 {+/-} 7.86 years) and 64 healthy controls (HC; mean age, 66.3 {+/-} 8.67) across three datasets: the Parkinsons Progression Markers Initiative (PPMI), the University of Pennsylvania (UPenn) and the Montreal Neurological Institute Biobank: Clinical Biological Imaging and Genetic Repository (C-BIG). The sample included different scanners, imaging parameters and lesion loads, as would be expected for multi-site data. WML were manually segmented to provide the gold standard, and four freely available automated algorithms were evaluated: FSLs BIANCA, FreeSurfer, SPMs LST-LPA and U-Net-pgs using the performance metrics: Dice score, Hausdorff distance, recall, precision, F1 score, log absolute volume difference (LOGAVD) and intraclass correlation coefficient (ICC). Subgroup analyses were performed based on lesion load and lobar regions. The associations of data from these automated approaches with age, and with Fazekas and Wahlund visual rating scales, were assessed through partial correlation analysis. ResultsU-Net-pgs performed best overall, with the highest Dice score (PD: 0.46 {+/-} 0.21; HC: 0.39 {+/-} 0.21), recall (PD: 0.76 {+/-} 0.25; HC: 0.62 {+/-} 0.31), precision (PD: 0.49 {+/-} 0.25; HC: 0.63 {+/-} 0.27), F1 score (PD: 0.54 {+/-} 0.22; HC: 0.56 {+/-} 0.22) and ICC (PD: 0.965; HC: 0.967) and lowest Hausdorff distance (PD: 8.89 {+/-} 3.96; HC: 6.33 {+/-} 2.91). U-Net-pgs achieved the lowest LOGAVD in the PD group (0.31 {+/-} 0.31) whereas BIANCA-LOO with a threshold of 0.9 was lowest in HC (0.27 {+/-} 0.30). U-Net also showed superior performances in all lesion loads for PD and overall across various brain regions in both PD and HC. ConclusionOverall, U-Net-pgs emerged as the best performing automated method, of those we evaluated, for WML segmentation in PD and HC within a dataset collected with various scanner and image acquisition parameters. U-Net-pgs consistently outperformed other automated approaches across lesion loads and brain regions, for most metrics. The accuracy and reliability of U-Net-pgs make it a promising tool for large-scale analyses, facilitating future research investigating WML in PD.

Reconstructing speech envelopes from electroen-cephalography (EEG) signals is a challenging but valuable task for brain-computer interfaces (BCIs), with applications in assistive communication for individuals with speech impairments. While deep learning has improved reconstruction accuracy, most existing approaches are restricted to single-layer architectures such as convolutional neural networks (CNNs). This limits their ability to capture the full complexity of spatio-temporal and structural EEG patterns. In this work, we systematically extend the VLAAI framework by evaluating 26 architectures that integrate CNNs, long short-term memory networks (LSTMs), and graph convolutional networks (GCNs) in both single-layer and hybrid configurations. Experiments on the 64-channel Spar-rKULee dataset demonstrate that CNNs remain the strongest standalone models, but hybrid designs--particularly CNN-LSTM and CNN-GCN-LSTM--achieve competitive or superior performance. These results highlight the importance of combining spatial, temporal, and graph-based processing, and provide practical guidelines for hybrid architecture design. Our study offers the first large-scale comparative analysis of hybrid models for EEG-based speech envelope reconstruction, advancing robust BCI systems for non-invasive speech decoding.

Understanding how the brain operates in naturalistic settings requires methods that go beyond conventional repeated-measurement approaches, necessitating the development of single-trial neural activity analysis. Recent advances in machine learning offer new opportunities for analyzing brain electrophysiological signals. Here, we recorded surface electrocorticography (ECoG) and intracranial local field potentials (LFPs) from emotion-related brain regions in a monkey performing a Pavlovian conditioning task, in which sensory cues predicting reward or punishment were presented randomly, followed by the actual unconditioned outcome. We evaluated the performance of two machine learning algorithms, a Convolutional neural network (CNN) model and a Transformer-based model (EEG-Conformer), in classifying raw ECoG/LFP traces. Both models successfully classified valence type during conditioned and unconditioned stimulus presentation. Furthermore, the Transformer achieved significantly superior classification performance compared to the CNN, particularly in multi-state classification including baseline periods. By optimizing the training dataset for the Transformer model, we could detect dynamic fluctuations in emotional valence consistent with task type from continuously evolving ECoG/LFP patterns recorded throughout the task. These results demonstrate the utility of Transformer-based models for decoding emotional valence from neurophysiological signals in non-human primates.

An important paradigm to study executive function is interval timing, which requires participants to estimate a temporal interval, often with a motor response. Interval timing translates from rodents to humans and can model neurodegenerative and neuropsychiatric disease. Interval timing also has parallel, time-dependent neurophysiology in rodents and humans, including time-dependent linear changes in neuronal firing rates over a temporal interval (ramping activity) and low-frequency [~]4 Hz "theta" activity evoked by trial start and response. Despite these translational features, an important confound of interval timing is movement and motor preparation, as participants must report their estimates by planning a movement. To address this confound, we first trained a group of 7 mice in a freely-moving interval timing task in which mice had to move across an operant chamber and nosepoke at the correct time to receive food reward. These mice were then trained in a head-fixed version of interval timing in which mice receive liquid reward for holding still for 3 seconds at the correct location, with reward access randomized to probe interval timing behavior. Despite vastly different motor sequences, we found prominent ramping activity in mouse prefrontal cortex ensembles and prefrontal cortical [~]4 Hz activity evoked both by trial start and preceding the timed decision. These data provide evidence that prefrontal neuronal ramping and theta activity is not linked to a specific motor program but rather a feature of the temporal organization of behavior.

Ethylene oxide is a widely used industrial chemical,yet evidence linking its exposure to Parkinsons disease remains limited.Using data from participants in the United States,we examined whether exposure to ethylene oxide is associated with Parkinson's disease.This cross-sectional study included 8,430 adults from the National Health and Nutrition Examination Survey (NHANES) collected between 2013 and 2020.Information on demographic characteristics,socioeconomic factors,lifestyle behaviors,body mass index,sedentary time and major chronic conditions was analyzed. Levels of hemoglobin ethylene oxide adducts,a biomarker of ethylene oxide exposure, were evaluated in relation to Parkinsons disease using statistical modeling approaches.After accounting for potential confounding factors,higher levels of ethylene oxide exposure were associated with an increased likelihood of Parkinson's disease.The association followed a positive and linear pattern.These findings provide new population-based evidence suggesting that ethylene oxide may be linked to Parkinsons disease and highlight the need for further studies to confirm causality and to better understand the biological mechanisms involved.

Traumatic brain injury (TBI) is a condition of high incidence worldwide, but remains mostly undertreated. Previous observations in preclinical studies pointed to a beneficial effect of insulin-like growth factor 1 (IGF-1) in TBI. As brain injury is associated to loss of IGF-1 sensitivity, we tested the therapeutic potential of AIK3a305 (AIK3), a novel IGF-1 sensitizer. Twenty-four hours after mild TBI induced by controlled impact, mice received daily intraperitoneal injections of AIK3 during 4 weeks. We found that TBI-associated sensorimotor disturbances measured with the adhesive-removal test were reverted by AIK3 treatment. In addition, neurological and cognitive disturbances measured by the neurological severity score and Y maze respectively, were also ameliorated by treatment with the IGF-1 sensitizer, whereas increased anxiety after mild TBI was also normalized by AIK3. Circulating levels of IGF-1 were increased after AIK3 treatment in TBI mice, while serum IL-6 levels, a biomarker of inflammation associated to TBI were similar to control mice treated with AIK3. Transcriptomic analysis determined that treatment with AIK3 widely affected gene expression in TBI brains, showing a general reduction in both up- and down-regulated genes. Collectively, these data support the use of IGF-1 sensitizers such as AIK3 for treatment of TBI.

BackgroundResting-state functional imaging is increasingly used to understand how cortical networks modulate and respond to pain. Awake imaging with a minimally invasive approach is key to observe the natural state of the brain. As migraine with aura, a common headache disorder, can be experimentally modeled by cortical spreading depressions (CSD) in rodents, it is essential to understand the impact of CSD on functional connectivity and network topology to find imaging cues of trigeminovascular activation and headache. MethodsWe used awake widefield intrinsic optical-signal imaging (IOSI) on optically cleared windows to non-invasively characterize the impact of CSDs on bihemispheric resting-state static and dynamic functional connectivity patterns and network topology. A subset of mice was chronically treated with amitriptyline to examine the effect of susceptibility to CSD on connectivity. After baseline imaging, CSD was triggered optogenetically and confirmed by laser speckle contrast imaging. A group of mice received intraperitoneal naproxen after CSD to suppress headache. IOSI was repeated at 30 minutes, 60 minutes, 4 hours, and 24 hours after CSD. The mouse grimace scale was scored at each time point for behavioral headache documentation. ResultsWe observed time-dependent changes in resting-state functional connectivity that were reversed by naproxen. Amitriptyline, a prophylactic migraine medication, decreased susceptibility to CSD and modified resting-state functional connectivity differently than controls. Network analysis with graph-theoretical methods revealed barrel and retrosplenial cortices as potential key players in trigeminal pain processing after CSD. Dynamic functional connectivity analysis demonstrated functional connectivity states, with fractional occupancy and mean dwell time of these states showing distinct CSD and pain-modulated states. A support vector machine was utilized to predict CSD-mediated dynamic connectivity changes in controls. ConclusionOur results bring insight into potentially headache-associated changes in resting-state cortical functional connectivity after CSD and how this functional reorganization is influenced by acute and chronic medications for migraine.