Bridging the behavior-neural gap: A multimodal AI reveals the brain's geometry of emotion more accurately than human self-reports

Bridging the behavior-neural gap: A multimodal AI reveals the brain's geometry of emotion more accurately than human self-reports | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Bridging the behavior-neural gap: A multimodal AI reveals the brain's geometry of emotion more accurately than human self-reports Huiguang He, Changde Du, Yizhuo Lu, Zhongyu Huang, Yi Sun, Zisen Zhou, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7545984/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract The ability to represent emotion plays a significant role in human cognition and social interaction, yet the high-dimensional geometry of this affective space and its neural underpinnings remain debated. A key challenge, the ‘behavior-neural gap,’ is the limited ability of human self-reports to predict brain activity. Here we test the hypothesis that this gap arises from the constraints of traditional rating scales and that large-scale similarity judgments can more faithfully capture the brain's affective geometry. Using AI models as ‘cognitive agents,’ we collected millions of triplet odd-one-out judgments from a multimodal large language model (MLLM) and a language-only model (LLM) in response to 2,180 emotionally evocative videos. We found that the emergent 30-dimensional embeddings from these models are highly interpretable and organize emotion primarily along categorical lines, yet in a blended fashion that incorporates dimensional properties. Most remarkably, the MLLM's representation predicted neural activity in human emotion-processing networks with the highest accuracy, outperforming not only the LLM but also, counterintuitively, representations derived directly from human behavioral ratings. This result supports our primary hypothesis and suggests that sensory grounding—learning from rich visual data—is critical for developing a truly neurally-aligned conceptual framework for emotion. Our findings provide compelling evidence that MLLMs can autonomously develop rich, neurally-aligned affective representations, offering a powerful paradigm to bridge the gap between subjective experience and its neural substrates. Project page: https://reedonepeck.github.io/ai-emotion.github.io/ Biological sciences/Neuroscience/Emotion Biological sciences/Computational biology and bioinformatics/Machine learning Biological sciences/Psychology/Human behaviour Biological sciences/Computational biology and bioinformatics/Computational neuroscience/Neural encoding Full Text Additional Declarations There is NO Competing Interest. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7545984","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":520679378,"identity":"dd3b108d-593e-4899-bc26-a20540f8e2e4","order_by":0,"name":"Huiguang He","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA4klEQVRIiWNgGAWjYBACAwbGBghLgoHxQQKpWpgNiNQCAxIMbBJEOcxcIrn5NU/NHbsG6eZjFQ9qGOT5G5ifPcCnxXJGYps1z7FnyQ0yx9JuJBxjMJxxgM3cAJ8WgxuJbcY8bIeTGSRyzG4ksDEwbmDgwe9CiJZ/IC353woS/jHYE6Ol+TFv22E7oC1sDIltDImEtZx52MY4t+9wAptEmrFEYp9E8ozDbGb4tRxPf/zhzbfD9vwSyQ8//vhmY9vf3vyMUGiDnQF0EhgA2cwE1AMB8wcgYU9Y3SgYBaNgFIxYAABodEXV0LBXeAAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-0684-1711","institution":"Institute of Automation, Chinese Academy of Sciences","correspondingAuthor":true,"prefix":"","firstName":"Huiguang","middleName":"","lastName":"He","suffix":""},{"id":520679379,"identity":"6b3d011e-bf72-4b43-b323-77b31433cd2e","order_by":1,"name":"Changde Du","email":"","orcid":"https://orcid.org/0000-0002-0084-433X","institution":"Institute of Automation, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Changde","middleName":"","lastName":"Du","suffix":""},{"id":520679380,"identity":"ff68da7f-d33c-4563-9fcb-9533ead631aa","order_by":2,"name":"Yizhuo Lu","email":"","orcid":"","institution":"Institute of Automation, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Yizhuo","middleName":"","lastName":"Lu","suffix":""},{"id":520679381,"identity":"1e94480e-a08b-450b-8fc6-c19d5662a348","order_by":3,"name":"Zhongyu Huang","email":"","orcid":"https://orcid.org/0000-0002-0085-5848","institution":"Institute of Automation, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Zhongyu","middleName":"","lastName":"Huang","suffix":""},{"id":520679382,"identity":"ad177e6c-67f7-488e-adef-3dda19b5da9d","order_by":4,"name":"Yi Sun","email":"","orcid":"https://orcid.org/0009-0005-3848-122X","institution":"Institute of Automation, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Yi","middleName":"","lastName":"Sun","suffix":""},{"id":520679383,"identity":"60aaa2d4-443b-4672-a8bd-bcdcd931746c","order_by":5,"name":"Zisen Zhou","email":"","orcid":"","institution":"Beijing Normal University","correspondingAuthor":false,"prefix":"","firstName":"Zisen","middleName":"","lastName":"Zhou","suffix":""},{"id":520679384,"identity":"2b2ed3d9-4458-4eb3-b034-29d5f7ba7a84","order_by":6,"name":"Shaozheng Qin","email":"","orcid":"https://orcid.org/0000-0002-1859-2150","institution":"State Key Laboratory of Cognitive Neuroscience and Learning \u0026 International Data Group/McGovern Institute for Brain Research, Beijing Normal University, Beijing","correspondingAuthor":false,"prefix":"","firstName":"Shaozheng","middleName":"","lastName":"Qin","suffix":""}],"badges":[],"createdAt":"2025-09-05 16:37:39","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7545984/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7545984/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":96240040,"identity":"8e8af6ac-37be-43f6-b61f-84c8c22517a9","added_by":"auto","created_at":"2025-11-19 07:08:17","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":24504573,"visible":true,"origin":"","legend":"Article File","description":"","filename":"main6small.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7545984/v1_covered_b66dd55f-a2df-4d37-b474-9cb9c4e6359d.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Bridging the behavior-neural gap: A multimodal AI reveals the brain's geometry of emotion more accurately than human self-reports","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7545984/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7545984/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"The ability to represent emotion plays a significant role in human cognition and social interaction, yet the high-dimensional geometry of this affective space and its neural underpinnings remain debated. A key challenge, the ‘behavior-neural gap,’ is the limited ability of human self-reports to predict brain activity. Here we test the hypothesis that this gap arises from the constraints of traditional rating scales and that large-scale similarity judgments can more faithfully capture the brain's affective geometry. Using AI models as ‘cognitive agents,’ we collected millions of triplet odd-one-out judgments from a multimodal large language model (MLLM) and a language-only model (LLM) in response to 2,180 emotionally evocative videos. We found that the emergent 30-dimensional embeddings from these models are highly interpretable and organize emotion primarily along categorical lines, yet in a blended fashion that incorporates dimensional properties. Most remarkably, the MLLM's representation predicted neural activity in human emotion-processing networks with the highest accuracy, outperforming not only the LLM but also, counterintuitively, representations derived directly from human behavioral ratings. This result supports our primary hypothesis and suggests that sensory grounding—learning from rich visual data—is critical for developing a truly neurally-aligned conceptual framework for emotion. Our findings provide compelling evidence that MLLMs can autonomously develop rich, neurally-aligned affective representations, offering a powerful paradigm to bridge the gap between subjective experience and its neural substrates. Project page: https://reedonepeck.github.io/ai-emotion.github.io/","manuscriptTitle":"Bridging the behavior-neural gap: A multimodal AI reveals the brain's geometry of emotion more accurately than human self-reports","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-26 10:28:57","doi":"10.21203/rs.3.rs-7545984/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"906dc785-feef-43d3-9c25-8d5320deba45","owner":[],"postedDate":"September 26th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":55348608,"name":"Biological sciences/Neuroscience/Emotion"},{"id":55348609,"name":"Biological sciences/Computational biology and bioinformatics/Machine learning"},{"id":55348610,"name":"Biological sciences/Psychology/Human behaviour"},{"id":55348611,"name":"Biological sciences/Computational biology and bioinformatics/Computational neuroscience/Neural encoding"}],"tags":[],"updatedAt":"2025-11-13T10:55:57+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-26 10:28:57","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7545984","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7545984","identity":"rs-7545984","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}