{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,26]],"date-time":"2026-10-26T07:35:44Z","timestamp":1793000144863,"version":"4.5.0"},"reference-count":45,"publisher":"IOP Publishing","issue":"5","license":[{"start":{"date-parts":[[2026,10,26]],"date-time":"2026-10-26T00:00:00Z","timestamp":1792972800000},"content-version":"vor","delay-in-days":25,"URL":"https:\/\/publishingsupport.iopscience.iop.org\/iop-standard\/v1"},{"start":{"date-parts":[[2026,10,26]],"date-time":"2026-10-26T00:00:00Z","timestamp":1792972800000},"content-version":"tdm","delay-in-days":25,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"name":"Natural Science Foundation of Beijing Municipality","award":["1232020"],"award-info":[{"award-number":["1232020"]}]},{"name":"Chinese Universities Scientific Fund","award":["2024JCYJ003"],"award-info":[{"award-number":["2024JCYJ003"]}]},{"name":"National Natural Science Foundation of China","award":["12371094"],"award-info":[{"award-number":["12371094"]}],"id":[{"id":"https:\/\/ror.org\/01h0zpd94","id-type":"ROR","asserted-by":"crossref"}]}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["J. Neural Eng."],"published-print":{"date-parts":[[2026,10,1]]},"abstract":"Abstract<\/jats:title>\n \n Objective<\/jats:italic>\n . Motor-imagery (MI) electroencephalography (EEG) offers a non-invasive route to brain\u2013computer interaction, yet reliable single-trial decoding remains difficult because task-related patterns are weak, nonstationary, and distributed over multiple temporal scales. This study investigates whether a staged local-to-global model can integrate fine temporal\u2013spatial structure with longer-range context without relying on unrestricted self-attention.\n Approach.<\/jats:italic>\n We propose MTDANet, an architecture that combines multi-scale temporal\u2013spatial feature extraction, factorized temporal consolidation, and distance-regularized recurrent attention. The model was evaluated on three public MI datasets under subject-specific and leave-one-subject-out (LOSO) protocols. Separately initialized models were also trained on P300 and steady-state visual evoked potential (SSVEP) datasets to examine broader architectural applicability. Participant-level paired analyses used two-sided paired\n \n \n \n <\/jats:tex-math>\n \n \n t<\/mml:mi>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n tests with within-setting Holm correction.\n Main results.<\/jats:italic>\n MTDANet achieved subject-specific accuracies of\n \n \n \n <\/jats:tex-math>\n \n \n 85.61<\/mml:mn>\n \u00b1<\/mml:mo>\n 8.32<\/mml:mn>\n %<\/mml:mi>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n ,\n \n \n \n <\/jats:tex-math>\n \n \n 90.85<\/mml:mn>\n \u00b1<\/mml:mo>\n 7.62<\/mml:mn>\n %<\/mml:mi>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n , and\n \n \n \n <\/jats:tex-math>\n \n \n 87.79<\/mml:mn>\n \u00b1<\/mml:mo>\n 13.93<\/mml:mn>\n %<\/mml:mi>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n on brain\u2013computer interface (BCI) Competition IV 2a, BCI Competition IV 2b, and BNCI2014_002, respectively. Relative to the highest-mean baseline, the paired improvements were supported after within-setting Holm correction on BCI Competition IV 2a (3.56 percentage points, adjusted\n \n \n \n <\/jats:tex-math>\n \n \n p<\/mml:mi>\n =<\/mml:mo>\n 0.0097<\/mml:mn>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n ) and 2b (1.62 points, adjusted\n \n \n \n <\/jats:tex-math>\n \n \n p<\/mml:mi>\n =<\/mml:mo>\n 0.0168<\/mml:mn>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n ). Under LOSO evaluation, the corresponding accuracies were\n \n \n \n <\/jats:tex-math>\n \n \n 63.31<\/mml:mn>\n \u00b1<\/mml:mo>\n 15.74<\/mml:mn>\n %<\/mml:mi>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n ,\n \n \n \n <\/jats:tex-math>\n \n \n 79.77<\/mml:mn>\n \u00b1<\/mml:mo>\n 7.57<\/mml:mn>\n %<\/mml:mi>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n , and\n \n \n \n <\/jats:tex-math>\n \n \n 79.96<\/mml:mn>\n \u00b1<\/mml:mo>\n 11.95<\/mml:mn>\n %<\/mml:mi>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n . The 2.08-point BNCI2014_002 difference had a 95% confidence interval of 0.42\u20133.73 points and an adjusted\n \n \n \n <\/jats:tex-math>\n \n \n p<\/mml:mi>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n value of 0.0387. These three comparisons did not remain significant in a broader 56-comparison Holm sensitivity analysis. The P300 and SSVEP models attained the highest observed mean accuracies.\n Significance.<\/jats:italic>\n The results provide within-setting evidence for staged local-to-global temporal modeling in offline MI-EEG evaluations.\n <\/jats:p>","DOI":"10.1088\/1741-2552\/aeb330","type":"journal-article","created":{"date-parts":[[2026,10,12]],"date-time":"2026-10-12T22:50:53Z","timestamp":1791845453000},"page":"056058","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["MTDANet: lightweight multi-scale temporal hierarchy and distance-regularized linear attention network for motor imagery brain\u2013computer interfaces"],"prefix":"10.1088","volume":"23","author":[{"ORCID":"https:\/\/orcid.org\/0009-0002-2815-9008","authenticated-orcid":true,"given":"Binbin","family":"Chen","sequence":"first","affiliation":[{"name":"Beijing Sport University"}],"role":[{"vocabulary":"crossref","role":"author"},{"role":"methodology","vocabulary":"credit"},{"role":"software","vocabulary":"credit"},{"role":"writing-original-draft","vocabulary":"credit"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0541-2069","authenticated-orcid":true,"given":"Chen","family":"Deng","sequence":"additional","affiliation":[{"name":"Beijing Sport University"}],"role":[{"vocabulary":"crossref","role":"author"},{"role":"data-curation","vocabulary":"credit"},{"role":"validation","vocabulary":"credit"},{"role":"visualization","vocabulary":"credit"}]},{"ORCID":"https:\/\/orcid.org\/0009-0005-8843-5221","authenticated-orcid":true,"given":"Huanmin","family":"Ge","sequence":"additional","affiliation":[{"name":"Beijing Sport University"}],"role":[{"vocabulary":"crossref","role":"author"},{"role":"conceptualization","vocabulary":"credit"},{"role":"funding-acquisition","vocabulary":"credit"},{"role":"supervision","vocabulary":"credit"},{"role":"writing-review-editing","vocabulary":"credit"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9156-9336","authenticated-orcid":false,"given":"Xinhua","family":"Su","sequence":"additional","affiliation":[{"name":"Beijing Sport University"}],"role":[{"vocabulary":"crossref","role":"author"},{"role":"formal-analysis","vocabulary":"credit"},{"role":"funding-acquisition","vocabulary":"credit"},{"role":"investigation","vocabulary":"credit"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8093-7754","authenticated-orcid":false,"given":"Yanfei","family":"Shen","sequence":"additional","affiliation":[{"name":"Beijing Sport University"}],"role":[{"vocabulary":"crossref","role":"author"},{"role":"project-administration","vocabulary":"credit"},{"role":"resources","vocabulary":"credit"}]}],"member":"266","published-online":{"date-parts":[[2026,10,26]]},"reference":[{"key":"jneaeb330bib1","doi-asserted-by":"publisher","first-page":"767","DOI":"10.1016\/S1388-2457(02)00057-3","type":"journal-article","article-title":"Brain\u2013computer interfaces for communication and control","volume":"113","author":"Wolpaw","year":"2002","journal-title":"Clin. Neurophysiol."},{"key":"jneaeb330bib2","author":"Wolpaw","year":"2012","type":"book"},{"key":"jneaeb330bib3","first-page":"pp 67","type":"book","article-title":"Brain\u2013computer interfaces","volume":"vol 110","author":"Wolpaw","year":"2013"},{"key":"jneaeb330bib4","doi-asserted-by":"publisher","first-page":"70","DOI":"10.1109\/TNSRE.2003.810426","type":"journal-article","article-title":"A general framework for brain-computer interface design","volume":"11","author":"Mason","year":"2003"},{"key":"jneaeb330bib5","doi-asserted-by":"publisher","first-page":"26","DOI":"10.1109\/MC.2012.107","type":"journal-article","article-title":"Brain-computer interfaces: Beyond medical applications","volume":"45","author":"van Erp","year":"2012","journal-title":"Computer"},{"key":"jneaeb330bib6","doi-asserted-by":"publisher","first-page":"153","DOI":"10.1016\/j.neuroimage.2005.12.003","type":"journal-article","article-title":"Mu rhythm (de)synchronization and EEG single-trial classification of different motor imagery tasks","volume":"31","author":"Pfurtscheller","year":"2006","journal-title":"NeuroImage"},{"key":"jneaeb330bib7","first-page":"pp 65","type":"book","article-title":"Electroencephalographic characteristics during motor imagery","author":"Neuper","year":"2010"},{"key":"jneaeb330bib8","doi-asserted-by":"publisher","first-page":"668","DOI":"10.1016\/j.cogbrainres.2005.08.014","type":"journal-article","article-title":"Imagery of motor actions: Differential effects of kinesthetic and visual-motor mode of imagery in single-trial EEG","volume":"25","author":"Neuper","year":"2005","journal-title":"Brain Res."},{"key":"jneaeb330bib9","doi-asserted-by":"publisher","first-page":"198","DOI":"10.3389\/fnins.2010.00198","type":"journal-article","article-title":"The Berlin brain-computer interface: Non-medical uses of BCI technology","volume":"4","author":"Blankertz","year":"2010","journal-title":"Front. Neurosci."},{"key":"jneaeb330bib10","doi-asserted-by":"publisher","DOI":"10.1088\/1741-2552\/aab2f2","type":"journal-article","article-title":"A review of classification algorithms for EEG-based brain\u2013computer interfaces: A 10 year update","volume":"15","author":"Lotte","year":"2018","journal-title":"J. Neural Eng."},{"key":"jneaeb330bib11","doi-asserted-by":"publisher","first-page":"1423","DOI":"10.3390\/s19061423","type":"journal-article","article-title":"EEG-based brain-computer interfaces using motor-imagery: Techniques and challenges","volume":"19","author":"Padfield","year":"2019","journal-title":"Sensors"},{"key":"jneaeb330bib12","doi-asserted-by":"publisher","first-page":"1425","DOI":"10.1109\/TBME.2014.2312397","type":"journal-article","article-title":"Brain\u2013computer interfaces using sensorimotor rhythms: Current state and future perspectives","volume":"61","author":"Yuan","year":"2014","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"jneaeb330bib13","doi-asserted-by":"publisher","first-page":"1303","DOI":"10.1016\/j.neuroimage.2010.03.022","type":"journal-article","article-title":"Neurophysiological predictor of SMR-based BCI performance","volume":"51","author":"Blankertz","year":"2010","journal-title":"NeuroImage"},{"key":"jneaeb330bib14","doi-asserted-by":"publisher","first-page":"65","DOI":"10.1016\/S0304-3940(97)00889-6","type":"journal-article","article-title":"Motor imagery activates primary sensorimotor area in humans","volume":"239","author":"Pfurtscheller","year":"1997","journal-title":"Neurosci. Lett."},{"key":"jneaeb330bib15","first-page":"pp 2390","type":"conference-proceedings","article-title":"Filter bank common spatial pattern (FBCSP)","author":"Ang","year":"2008"},{"key":"jneaeb330bib16","doi-asserted-by":"publisher","first-page":"5391","DOI":"10.1002\/hbm.23730","type":"journal-article","article-title":"Deep learning with convolutional neural networks for EEG decoding and visualization","volume":"38","author":"Schirrmeister","year":"2017","journal-title":"Hum. Brain Mapp."},{"key":"jneaeb330bib17","doi-asserted-by":"publisher","DOI":"10.1088\/1741-2552\/aace8c","type":"journal-article","article-title":"EEGNet: A compact convolutional neural network for EEG-based brain\u2013computer interfaces","volume":"15","author":"Lawhern","year":"2018","journal-title":"J. Neural Eng."},{"key":"jneaeb330bib18","article-title":"FBCNet: A multi-view convolutional neural network for brain-computer interface","author":"Mane","year":"2021","type":"preprint"},{"key":"jneaeb330bib19","doi-asserted-by":"publisher","first-page":"1900","DOI":"10.1109\/TNSRE.2023.3257319","type":"journal-article","article-title":"IFNet: An interactive frequency convolutional neural network for enhancing motor imagery decoding from EEG","volume":"31","author":"Wang","year":"2023"},{"key":"jneaeb330bib20","doi-asserted-by":"publisher","first-page":"710","DOI":"10.1109\/TNSRE.2022.3230250","type":"journal-article","article-title":"EEG Conformer: Convolutional Transformer for EEG decoding and visualization","volume":"31","author":"Song","year":"2023","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng"},{"key":"jneaeb330bib21","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-024-71118-7","type":"journal-article","article-title":"CTNet: A convolutional Transformer network for EEG-based motor imagery classification","volume":"14","author":"Zhao","year":"2024","journal-title":"Sci. Rep."},{"key":"jneaeb330bib22","doi-asserted-by":"publisher","first-page":"4134","DOI":"10.1109\/JBHI.2025.3622725","type":"journal-article","article-title":"DBConformer: Dual-branch convolutional Transformer for EEG decoding","volume":"30","author":"Wang","year":"2026","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"jneaeb330bib23","first-page":"pp 5998","type":"conference-proceedings","article-title":"Attention is all you need","volume":"vol 30","author":"Vaswani","year":"2017"},{"key":"jneaeb330bib24","first-page":"pp 5156","type":"conference-proceedings","article-title":"Transformers are RNNs: Fast autoregressive transformers with linear attention","author":"Katharopoulos","year":"2020"},{"key":"jneaeb330bib25","article-title":"Efficiently modeling long sequences with structured state spaces","author":"Gu","year":"2022","type":"conference-proceedings"},{"key":"jneaeb330bib26","article-title":"Mamba: Linear-time sequence modeling with selective state spaces","author":"Gu","year":"2023","type":"preprint"},{"key":"jneaeb330bib27","first-page":"pp 14048","type":"book","article-title":"RWKV: Reinventing RNNs for the transformer era","author":"Peng","year":"2023"},{"key":"jneaeb330bib28","doi-asserted-by":"publisher","first-page":"154","DOI":"10.1109\/TNSRE.2023.3342331","type":"journal-article","article-title":"ADFCNN: Attention-based dual-scale fusion convolutional neural network for motor imagery brain-computer interface","volume":"32","author":"Tao","year":"2023"},{"key":"jneaeb330bib29","article-title":"Retentive network: A successor to transformer for large language models","author":"Sun","year":"2023","type":"preprint"},{"key":"jneaeb330bib30","doi-asserted-by":"publisher","DOI":"10.3389\/fnins.2024.1366294","type":"journal-article","article-title":"EMPT: A sparsity Transformer for EEG-based motor imagery recognition","volume":"18","author":"Liu","year":"2024","journal-title":"Front. Neurosci."},{"key":"jneaeb330bib31","author":"Brunner","year":"2008","type":"other"},{"key":"jneaeb330bib32","article-title":"BCI competition 2008 \u2013 Graz data set B","author":"Leeb","year":"2008","type":"other"},{"key":"jneaeb330bib33","doi-asserted-by":"publisher","first-page":"77","DOI":"10.1515\/bmt-2014-0117","type":"journal-article","article-title":"Random forests in non-invasive sensorimotor rhythm brain\u2013computer interfaces: a practical and convenient non-linear classifier","volume":"61","author":"Steyrl","year":"2016","journal-title":"Biomed. Tech."},{"key":"jneaeb330bib34","article-title":"Covert and overt ERP-based BCI BNCI Horizon 2020, dataset 009-2014","year":"2026","type":"other"},{"key":"jneaeb330bib35","doi-asserted-by":"publisher","first-page":"55","DOI":"10.1016\/j.neucom.2016.01.007","type":"journal-article","article-title":"Online SSVEP-based BCI using Riemannian geometry","volume":"191","author":"Kalunga","year":"2016","journal-title":"Neurocomputing"},{"key":"jneaeb330bib36","doi-asserted-by":"publisher","first-page":"2773","DOI":"10.1109\/TNSRE.2020.3048106","type":"journal-article","article-title":"EEG-Inception: A novel deep convolutional neural network for assistive ERP-based brain\u2013computer interfaces","volume":"28","author":"Santamar\u00eda-V\u00e1zquez","year":"2020"},{"key":"jneaeb330bib37","doi-asserted-by":"publisher","first-page":"2238","DOI":"10.1109\/TAFFC.2022.3169001","type":"journal-article","article-title":"TSception: Capturing temporal dynamics and spatial asymmetry from EEG for emotion recognition","volume":"14","author":"Ding","year":"2022","journal-title":"IEEE Trans. Affect. Comput."},{"key":"jneaeb330bib38","doi-asserted-by":"publisher","DOI":"10.1088\/1741-2552\/aadea0","type":"journal-article","article-title":"MOABB: Trustworthy algorithm benchmarking for BCIs","volume":"15","author":"Jayaram","year":"2018","journal-title":"J. Neural Eng."},{"key":"jneaeb330bib39","doi-asserted-by":"publisher","first-page":"55","DOI":"10.3389\/fnins.2012.00055","type":"journal-article","article-title":"Review of the BCI competition IV","volume":"6","author":"Tangermann","year":"2012","journal-title":"Front. Neurosci."},{"key":"jneaeb330bib40","doi-asserted-by":"publisher","DOI":"10.1088\/1741-2560\/11\/3\/035008","type":"journal-article","article-title":"Influence of P300 latency jitter on event related potential-based brain\u2013computer interface performance","volume":"11","author":"Aric\u00f2","year":"2014","journal-title":"J. Neural Eng."},{"key":"jneaeb330bib41","first-page":"871","type":"conference-proceedings","article-title":"Signal processing approaches to minimize or suppress calibration time in oscillatory activity-based brain\u2013computer interfaces","author":"Lotte","year":"2015"},{"key":"jneaeb330bib42","first-page":"pp 80233","type":"conference-proceedings","article-title":"Mitigating over-smoothing in transformers via regularized nonlocal functionals","volume":"vol 36","author":"Nguyen","year":"2023"},{"key":"jneaeb330bib43","article-title":"Revisiting over-smoothing in BERT from the perspective of graph","author":"Shi","year":"2022","type":"conference-proceedings"},{"key":"jneaeb330bib44","doi-asserted-by":"publisher","first-page":"96","DOI":"10.1016\/j.neuroimage.2013.10.067","type":"journal-article","article-title":"On the interpretation of weight vectors of linear models in multivariate neuroimaging","volume":"87","author":"Haufe","year":"2014","journal-title":"NeuroImage"},{"key":"jneaeb330bib45","first-page":"2579","type":"journal-article","article-title":"Visualizing data using t-SNE","volume":"9","author":"Van der Maaten","year":"2008","journal-title":"J. Mach. Learn. Res."}],"container-title":["Journal of Neural Engineering"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/aeb330","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/aeb330\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/aeb330","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/aeb330\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/aeb330\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/aeb330\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/aeb330\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/aeb330\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,26]],"date-time":"2026-10-26T07:22:53Z","timestamp":1792999373000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/aeb330"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10,1]]},"references-count":45,"journal-issue":{"issue":"5","published-online":{"date-parts":[[2026,10,26]]},"published-print":{"date-parts":[[2026,10,1]]}},"URL":"https:\/\/doi.org\/10.1088\/1741-2552\/aeb330","relation":{},"ISSN":["1741-2560","1741-2552"],"issn-type":[{"value":"1741-2560","type":"print"},{"value":"1741-2552","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,1]]},"assertion":[{"value":"MTDANet: lightweight multi-scale temporal hierarchy and distance-regularized linear attention network for motor imagery brain\u2013computer interfaces","name":"article_title","label":"Article Title"},{"value":"Journal of Neural Engineering","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2026 IOP Publishing Ltd. All rights, including for text and data mining, AI training, and similar technologies, are reserved.","name":"copyright_information","label":"Copyright Information"},{"value":"2026-06-15","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2026-10-12","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2026-10-26","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}