Journal of neural engineering最新文献

{"title":"The calibration-to-accuracy mapping in P300 spellers does not transport across cohorts and was not restored by tested alignment or local recalibration.","authors":"Alon Gorenshtein, Yosef Adiniaev, Tom Liba, Yiftach Barash, Eyal Klang, Oved Daniel","doi":"10.1088/1741-2552/aeb62c","DOIUrl":"https://doi.org/10.1088/1741-2552/aeb62c","url":null,"abstract":"<p><p>Objective&#xD;A calibration-derived score has been related to online P300-speller accuracy in the same session, always within the cohort measured. We evaluated whether a mapping from that score to expected accuracy transports to withheld cohorts.&#xD;Approach&#xD;Retrospective secondary analysis of BigP3BCI 1.0.0, the largest evaluation of this relationship to date: 18 of 20 studies contributed 271 participants, 739 session-condition records, and 19,611 character selections. The predictor was a decoder's cross-validated discriminability on calibration epochs, one study withheld from development. Between-cohort variation was summarised by the random-effects standard deviation tau.&#xD;Main Results&#xD;The association was positive in all 18 cohorts but varied widely (r 0.190 to 0.928; participant-level r = 0.714, p < 0.001), with 12 CIs excluding zero. Pooled estimation error was 0.098 (95% CI 0.091 to 0.107) against a benchmark of 0.146. Calibration did not transport: intercept tau 0.87 (unrepresented-cohort interval -1.97 to 1.85, log-odds); cohort-specific slopes varied more than tenfold (tau 0.43, interval 0.111 to 2.005). Neither alignment (signal- or score-space) nor either nonlinear boundary meaningfully reduced this heterogeneity. Local recalibration with up to 16 participants did not reliably improve on transporting; refitting both intercept and slope was significantly worse than transporting at the smallest sizes. Median time per selection, a stopping-rule proxy, accounted for 66% of between-cohort slope variance in meta-regression. The four ALS cohorts gave an uncorrected slope spread of 0.223, against 0.560 overall.&#xD;Significance&#xD;The evaluated score is a classifier's cross-validated discriminability on each session's calibration data rather than the source studies' undocumented online decoder. That score reproducibly signals within-session accuracy, but its mapping to accuracy is cohort-specific and supports ranking sessions after local validation rather than reporting accuracy elsewhere. The four-cohort ALS subgroup appeared more homogeneous than the full archive, illustrating how conclusions about transportability from a small cohort set can be fragile.&#xD.</p>","PeriodicalId":94096,"journal":{"name":"Journal of neural engineering","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2026-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149225745","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessing consciousness by near-nerve electrostimulation in intensive rehabilitation.","authors":"Piergiuseppe Liuzzi, Sara Ballanti, Andrea Mannini, Bahia Hakiki, Rachele Burali, Tanita Toci, Maria Chiara Carrozza, Calogero Maria Oddo, Antonello Grippo","doi":"10.1088/1741-2552/aea645","DOIUrl":"10.1088/1741-2552/aea645","url":null,"abstract":"<p><p><i>Objective.</i>Consciousness diagnosis in prolonged disorders of consciousness (pDoC) relies on the Coma Recovery Scale-Revised, yet consciousness or its potential for recovery may escape behavioural detection. We tested whether near-nerve stimulation (NNS) of the median nerve, compared to standard surface cutaneous stimulation (SCS), could provide an EEG-based marker of consciousness and its recovery, by specifically eliciting sensory responses.<i>Approach.</i>Nineteen pDoC patients (median age 69 years, time post-onset 74 d) underwent NNS and SCS with high-density EEG. We quantified evoked potentials, global field power and perturbational complexity, and defined recovery as improvement in clinical diagnosis over 2 months.<i>Main results.</i>A combined NNS index (field-power peak and perturbational complexity index) classified chronic unresponsive patients with a leave-one-subject-out accuracy of 94.7% (95% CI 77.9%-99.4%), sensitivity 100% (95% CI 73.8%-100%) and specificity 90.9% (95% CI 64.7%-99.0%), outperforming SCS (accuracy 63.2%).<i>Significance.</i>These findings suggest that NNS, by selectively activating sensory pathways without intentional motor recruitment, is a promising bedside tool. These are internal cross-validated estimates from a single-centre pilot cohort; external validation is required.</p>","PeriodicalId":94096,"journal":{"name":"Journal of neural engineering","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2026-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148946427","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Mapping unintended trigeminal nerve activation during transcranial alternating current stimulation (tACS): a computational study.","authors":"Sulagna Sahu, Munish Chauhan, Saurav Z K Sajib, Enock Boakye, Rosalind J Sadleir","doi":"10.1088/1741-2552/aeb074","DOIUrl":"10.1088/1741-2552/aeb074","url":null,"abstract":"<p><p><i>Objective.</i>To systematically characterize activation thresholds for peripheral trigeminal nerve branches under clinical transcranial alternating current stimulation (tACS) parameters using a computational modeling approach.<i>Approach.</i>An MRI-derived finite element head model was coupled with multicompartment NEURON cable models of myelinated sensory axons (4-14<i>µ</i>m diameter). Seven trigeminal branches were manually delineated. Activation thresholds were computed across eight sinusoidal frequencies (5-100 Hz) for two electrode montages (F3-F4 and Fp1-Fp2). A 60 Hz biphasic pulse enabled direct comparison with established transcutaneous nerve stimulation devices. Sensitivity analyses assessed robustness to nerve orientation and depth, and electrode displacement.<i>Main Results.</i>Threshold decreased non-linearly with increasing frequency, plateauing above 40 Hz. Under F3-F4, the auriculotemporal and zygomaticotemporal nerves, and under Fp1-Fp2, the supraorbital and supratrochlear nerves fell below 2 mA threshold at ⩾40 Hz. Thresholds were largely independent of axon diameter above 40 Hz. Thresholds for 60 Hz sinusoidal waveforms were consistently lower than 60 Hz biphasic pulse condition. Sensitivity analyses confirmed sub-2 mA thresholds were preserved for tested perturbations.<i>Significance.</i>Select trigeminal branches are activated below the standard 2 mA clinical dose across both montages. This study provides preliminary indication that peripheral trigeminal recruitment is a predictable, montage-dependent, and mechanistically plausible contributor to tACS effects that warrants explicit consideration in trial design and result interpretation.</p>","PeriodicalId":94096,"journal":{"name":"Journal of neural engineering","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2026-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149120962","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Involuntary facial muscle activity during imagined vocalisation contaminates EEG and enables emotion decoding.","authors":"Yichen Tang, Paul M Corballis, Luke E Hallum","doi":"10.1088/1741-2552/aeb073","DOIUrl":"10.1088/1741-2552/aeb073","url":null,"abstract":"<p><p><i>Objective.</i>Decoding imagined speech from electroencephalography (EEG) recordings is potentially useful for brain-computer interfaces. Previous studies have focused on decoding semantic information from EEG, leaving the decoding of emotion - an important component of human communication - largely unexplored.<i>Approach.</i>Here, we report two experiments involving participants tasked with overt (<i>n</i>= 14) or imagined (<i>n</i>= 21) emotional vocalisation in five different categories: anger, happiness, neutral, sadness, and pleasure. Throughout, we recorded 64-channel EEG; we computed time-frequency features and used a logistic-regression classifier to evaluate emotion decoding accuracy. In five participants, we also recorded facial surface electromyography (sEMG) during imagined vocalisation, and studied the contamination of EEG by sEMG.<i>Main results.</i>Our results show that emotion can be decoded from single-trial EEG recordings of both overt (78.1%, chance = 20%) and imagined vocalisation (36.4%). The high-gamma band (50 to 100 Hz) and lateral EEG channels (T7, T8, and proximal) were important for decoding. sEMG analysis indicated that involuntary facial muscle activity contributed to these spectral and spatial patterns during imagined vocalisation. We identified a prominent muscle-induced pattern over lateral EEG channels - the railroad cross-tie pattern - showing greater activation during imagined happy vocalisations. We conclude that involuntary facial muscle activity is associated with certain emotion categories (i.e. happiness) and contributes to the decoding of emotion from single-trial EEG recordings.<i>Significance.</i>This study provides converging EEG and sEMG evidence demonstrating that involuntary facial muscle activity occurs during imagined vocalisation and is modulated by emotion, producing emotion-specific EEG contamination patterns. Although we focused on imagined vocalisations, the EMG patterns we observed closely resembled EEG patterns commonly reported in emotion decoding studies, raising concerns about involuntary facial muscle activity in such work.</p>","PeriodicalId":94096,"journal":{"name":"Journal of neural engineering","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2026-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149120972","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Prioritized nonlinear modeling of shared dynamics across neural populations.","authors":"Trisha Jha, Dong Song, Maryam M Shanechi","doi":"10.1088/1741-2552/ae9bee","DOIUrl":"10.1088/1741-2552/ae9bee","url":null,"abstract":"<p><p><i>Objective</i>. Advances in neural recording technology enable simultaneous measurements across multiple brain regions, providing new opportunities to study inter-regional interactions. However, several challenges arise when developing nonlinear dynamical models of cross-region interactions. In particular, we identify three key desired properties for a nonlinear framework. First, it should prioritize extraction of cross-region dynamics to avoid confounds from within-region dynamics. Second, it should enable localization of nonlinear structure within the model to better interpret cross-regional interaction models. Third, it should support both temporally causal and non-causal inference of shared dynamics and do so using source region activity alone. Current cross-regional models do not satisfy all these properties.<i>Approach</i>. Here, we address these challenges by developing cross-population prioritized dynamical nonlinear interaction model (CroP-DYNO), a nonlinear dynamical framework that prioritizes the learning of shared cross-population dynamics to avoid confounds from within-population activity. Further, CroP-DYNO allows individual components of the dynamical model to be independently specified as nonlinear or linear. Finally, it supports both temporally causal and non-causal inference of latent states, using only source region activity.<i>Main results</i>. We validate our method on datasets across species and distinct brain regions. We find that both the prioritized learning and the nonlinear modeling in CroP-DYNO are important for accurately extracting cross-population dynamics. As such, CroP-DYNO outperforms baseline nonprioritized and linear prioritized models in predicting target neural population activity from source activity. Further, CroP-DYNO enables systematic localization of nonlinear structure and quantifies dominant predictive interactions between brain regions, with interaction strengths that align with known circuit anatomy.<i>Significance</i>. Overall, these results establish CroP-DYNO as a flexible nonlinear framework for studying interactions between neural populations and brain regions, enabling both accurate modeling and interpretable dissection of nonlinear structure in cross-regional communication models.</p>","PeriodicalId":94096,"journal":{"name":"Journal of neural engineering","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2026-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13673170/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148803480","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Phase-locked closed-loop ultrasound stimulation of hippocampal slow gamma modulates CA3-CA1 coherence and improves spatial memory in APP/PS1 mice.","authors":"Zhenyu Xie, Hui Ji, Mürsel Karadas, Yiyao Zhang, Yi Yuan","doi":"10.1088/1741-2552/aeb075","DOIUrl":"10.1088/1741-2552/aeb075","url":null,"abstract":"<p><p><i>Objective.</i>Impaired hippocampal slow gamma oscillations and CA3-CA1 dysfunction may contribute to spatial memory deficits in early Alzheimer's disease (AD). This study examined whether phase-locked closed-loop transcranial ultrasound stimulation (TUS) could modulate CA3-CA1 activity and improve memory.<i>Approach.</i>A closed-loop TUS system was developed that uses the phase of endogenous slow gamma (30-45 Hz) rhythms in CA3 local field potentials as a trigger. Acute experiments were performed to assess phase-dependent neural responses. Normal rats received 14 d of stimulation to examine circuit connectivity and behavior. APP/PS1 mice received 14 d of peak-phase stimulation to test behavioral and electrophysiological effects.<i>Main results.</i>Acute experiments showed that both stimulation phase and stimulation number jointly modulated neural responses. Peak-phase and random-phase stimulation increased slow gamma power and cross-frequency coupling, while peak-phase stimulation also decreased sample entropy and increased CA3-CA1 coherence. In normal rats, behavioral performance did not differ among groups after 14 d, but task-related CA3-CA1 slow gamma coherence was higher in the peak-phase group. In APP/PS1 mice, peak-phase stimulation improved Y-maze spontaneous alternation and novel object location discrimination, alongside increased CA1 power, CA3-CA1 coherence, cross-frequency coupling, and spike-field coherence, all of which were positively correlated with behavioral scores.<i>Significance.</i>These findings indicate that phase-locked closed-loop ultrasound stimulation modulates hippocampal slow gamma activity and CA3-CA1 connectivity in a phase-dependent manner and improves spatial memory in an AD mouse model, supporting oscillation-guided noninvasive neuromodulation as a potential intervention strategy.</p>","PeriodicalId":94096,"journal":{"name":"Journal of neural engineering","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2026-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149120942","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Single-channel EEG denoising and artifact removal: a systematic scoping review of methods, benchmarking, and reproducibility.","authors":"Usman Qamar Shaikh, Anubha Manju Kalra, Andrew Lowe, Imran Khan Niazi","doi":"10.1088/1741-2552/aeb0ce","DOIUrl":"10.1088/1741-2552/aeb0ce","url":null,"abstract":"<p><p><i>Objective.</i>Single-channel and wearable electroencephalography (EEG) reduce acquisition burden but limit the spatial information available to conventional multichannel artifact-removal methods. We conducted a systematic scoping review of single-channel-compatible EEG artifact removal, linking method operation with artifact coverage, evaluation design, generalization, deployment, comparison, and reproducibility.<i>Approach.</i>Database and citation searches through 15 2026 identified 178 articles describing 199 method variants. For learned methods, 303 reported model-dataset evaluations were also examined for data independence, generalization, and deployment.<i>Main results.</i>Most variants were reference-free at inference (179/199), while 13 used a non-neural auxiliary reference. Methods spanned seven operating paradigms, including decomposition, transformed source separation, adaptive correction, and learned reconstruction. Ocular and muscle artifacts were evaluated most often, and 38/178 articles tested simultaneous or overlapping mixtures. Across 380 dataset uses, evaluation centered on controlled signal reconstruction. Known-target fidelity was reported for 170/199 variants, compared with 21 for clean-input or non-artifact signal comparison, 32 for contaminated downstream utility, and three for clean-input downstream non-degradation. Among 81 learned variants, 18 reported source-trained application to an external dataset, device, or site; only 31/303 evaluations explicitly separated test windows or epochs, clean sources, and artifact sources from training. No measured deployment evidence was reported for 122/199 variants. Public method resources were available for 21 variants, while trained weights and explicit licenses were uncommon.<i>Significance.</i>The review provides a framework for selecting datasets, metrics, comparators, and implementations, and for distinguishing reconstruction, neural preservation, transfer, and wearable deployment evidence.</p>","PeriodicalId":94096,"journal":{"name":"Journal of neural engineering","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2026-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149120960","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Peripheral physiological decoding of arousal in a boundary-avoidance task: signal validity and individualized control.","authors":"Anirudh Meyyappan Natarajan, Paul Sajda","doi":"10.1088/1741-2552/aeb4ec","DOIUrl":"https://doi.org/10.1088/1741-2552/aeb4ec","url":null,"abstract":"<p><strong>Objective: </strong>Scoring well on a task-demand label does not show that a decoder measures arousal. We ask whether peripheral physiology decodes demand better than EEG, whether the index behaves like graded arousal rather than a clock, and what a controller could do with it.&#xD;Approach. We re-analysed a public boundary-avoidance flight task in which sixteen participants received EEG neurofeedback, half-sham feedback or silence. A decoder trained on peripheral signals (cardiac, respiratory, electrodermal, pupillometric) from an open-loop calibration block was compared with a retrained EEG decoder, filter-bank common spatial patterns, and motor-only and shuffled-label controls. All analyses are offline; the decoder was never in the loop.&#xD;Main results. The peripheral decoder reached 93.0% within-subject AUC against 85.2% for EEG. Ring size shrinks on a fixed schedule, so accuracy alone cannot separate arousal from trial position. Three tests did: 48.6% of hard-course trials ended below the scale midpoint; the index separated the courses at matched elapsed time (p= 1.4 ×10-31), where EEG did not; and it tracked the autonomic channels. Only the peripheral index recovered an inverted-U with performance, robust in a duration-safe hazard analysis (p= 3.5 ×10-10). Arousal rose 14.1 points in the 10 s before failure; sustained between-condition differences above 7.5 points are excluded. Subjects varied reliably in how strongly excess arousal predicted failure, predictably from pre-loop calibration. Personalised control bands gave the best held-out discrimination, though at N = 16 indistinguishable from a retuned universal band.&#xD;Significance. Arousal is a state of the whole body, and here the body reads it better than the scalp. The results separate three often-conflated requirements: accuracy on a demand label, validity as an arousal signal, and usefulness for control. The signal that added information over a fixed rule was distance above a band set by each subject's recent variability.</p>","PeriodicalId":94096,"journal":{"name":"Journal of neural engineering","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2026-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149198497","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Cybertissue Neuroprosthetics: a new implantable neuroprosthetics paradigm to achieve seamless lifetime integration.","authors":"Liang Guo","doi":"10.1088/1741-2552/aeb442","DOIUrl":"https://doi.org/10.1088/1741-2552/aeb442","url":null,"abstract":"<p><p>Despite decades of advances in neural interface technology, the long-term instability of implantable neuroprostheses remains a major barrier to achieving reliable lifetime performance. Chronic foreign-body reactions and fibrotic encapsulation progressively degrade the electrode-tissue interface, leading to signal deterioration and eventual device failure. Existing strategies that focus on improving material biocompatibility, reducing mechanical mismatch, or modifying implant surfaces have produced important but largely incremental improvements, while leaving the fundamental challenge unresolved. This perspective paper introduces a fundamentally different paradigm: Cybertissue Neuroprosthetics (CN). CN moves beyond the traditional electrode‑based neural interface by treating the implant itself as a functional tissue device-composed of living cells and optionally augmented with minimal nanoelectronics, nanomaterials, or genetic modifications-designed to integrate structurally and functionally with the host nervous system to achieve long‑term neuromodulation and information exchange. By minimizing or eliminating foreign materials and leveraging the brain's own cellular signaling mechanisms, CN promises to overcome the inherent limitations of current inorganic neuroprosthetic devices and seeks to achieve seamless lifetime integration between neuroprosthetic devices and the nervous system. This paper introduces the CN concept, unifies a range of emerging research directions under this umbrella, and articulates the core opportunities-from engineered cybercells to transplantable living devices-that collectively define a new interdisciplinary frontier. We argue that CN represents a foundational shift in neuroprosthetics, with the potential to deliver truly lifetime neural implants that are biologically integrated, self‑sustaining, and clinically transformative.</p>","PeriodicalId":94096,"journal":{"name":"Journal of neural engineering","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2026-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149187023","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Explainable fMRI decoding reveals neurobiological signatures of emotional arousal complementary to traditional analyses.","authors":"Daniel Agostinho, Davide Borra, Miguel Castelo-Branco, Marco Simões","doi":"10.1088/1741-2552/aead27","DOIUrl":"10.1088/1741-2552/aead27","url":null,"abstract":"<p><p><i>Objective.</i>Understanding how emotional arousal is represented in the brain is fundamental for elucidating affective regulation and its alterations in autism spectrum disorder. Classical neuroimaging approaches, such as the general linear model (GLM), have provided important insights into emotional processing. However, the extent to which data-driven decoding and traditional model-based analyses provide complementary information about emotional arousal remains unclear.<i>Approach.</i>We introduce an explainable artificial intelligence framework that combines the lightweight functional magnetic resonance imaging (fMRI) decoder (fMRINet) with Deep Learning Important FeaTures to derive data-driven neurobiological signatures of emotional arousal. The framework enables interpretation of spatiotemporal neural contributions during emotional processing. We applied the framework to fMRI data acquired while neurotypically and autistic participants viewed naturalistic emotional videos and compared the resulting signatures with categorical and parametric GLM analyses.<i>Main results.</i>The framework identified distinct spatiotemporal neurobiological signatures associated with emotional arousal. Neurotypical participants engaged a distributed perceptual-limbic-prefrontal network under high arousal and frontoparietal executive regions under low arousal. In contrast, autistic participants exhibited a more spatially restricted and posterior-biased pattern under high arousal, with reduced ventromedial prefrontal and limbic contributions, while low-arousal patterns resembled typically developing, suggesting preserved executive mechanisms during low-arousal processing. The GLM analyses revealed partly convergent and partly distinct findings: categorical GLM detected no significant group differences, whereas parametric GLM identified widespread cortical-subcortical modulation in neurotypical individuals and cerebellar-restricted effects in autistic, consistent with compensatory reliance on motor-timing circuits.<i>Significance</i>. The proposed framework provides an interpretable, data-driven perspective on emotional processing and reveals neurobiologically plausible patterns consistent with existing literature. Together, the explainable decoding and GLM results highlight complementary aspects of emotional arousal processing, suggesting that explainable decoding approaches may provide additional insights alongside traditional model-based neuroimaging analyses.</p>","PeriodicalId":94096,"journal":{"name":"Journal of neural engineering","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2026-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149077233","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}