Active inference as a unified model of collision avoidance behavior in human drivers

Active inference as a unified model of collision avoidance behavior in human drivers | 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 Active inference as a unified model of collision avoidance behavior in human drivers Julian Schumann, Johan Engstrom, Leif Johnson, Matthew O'Kelly, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6804690/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Collision avoidance -- involving a rapid threat detection and quick execution of the appropriate evasive maneuver -- is a critical aspect of driving. However, existing models of human collision avoidance behavior are fragmented, focusing on specific scenarios or only describing certain aspects of the avoidance behavior, such as response times. This paper addresses these gaps by proposing a novel computational cognitive model of human collision avoidance behavior based on active inference. Active inference provides a unified approach to modeling human behavior: the minimization of free energy. Building on prior active inference work, our model incorporates established cognitive mechanisms such as evidence accumulation to simulate human responses in two distinct collision avoidance scenarios: front-to-rear lead vehicle braking and lateral incursion by an oncoming vehicle. We demonstrate that our model explains a wide range of previous empirical findings on human collision avoidance behavior. Specifically, the model closely reproduces both aggregate results from meta-analyses previously reported in the literature and detailed, scenario-specific effects observed in a recent driving simulator study, including response timing, maneuver selection, and execution. Our results highlight the potential of active inference as a unified framework for understanding and modeling human behavior in complex real-life driving tasks. Physical sciences/Engineering/Civil engineering Physical sciences/Engineering/Mechanical engineering Full Text Additional Declarations There is NO Competing Interest. Supplementary Files Simulationfile3A.mp4 Simulation underlying Figure 3A Simulationfile3B.mp4 Simulation underlying Figure 3B ActiveInferenceSupplementarydescriptions.pdf Description of supplementary files ActiveInferenceSupplementarymaterials.pdf LaTeX Supplementary File Simulationfile4A.mp4 Simulation underlying Figure 4A Simulationfile4B.mp4 Simulation underlying Figure 4B NoEAshallowincurisoncollision.mp4 Video referenced in the article Cite Share Download PDF Status: Under Review 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-6804690","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":501304950,"identity":"c436f448-9f8b-4f7c-96d6-030cbed68c9f","order_by":0,"name":"Julian Schumann","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA7ElEQVRIie3RoQrCQBzH8f+fCyvTYVPGHkIGQ1DfZVe0DJ/AcCBoEVbnQwimYXNysDRZNRgUwWSY7aInimLYuWi4b7odfPjBDUCn+8MMZhUoIHl+FQDIDDzKY6eUIDOBmC+CkS8vCGkzgKaawIsQszoZH+hytkvPfcGdkBDjiGsVsRIU6YXG2WjoBj53FxOCDDP1CtQYp3ESeLYkdMWtLcNpFZJfPbsjyYY/VhSk8SZ7uQKPFfKD2PKFuZlyN95fvdZ8MHQjuRJRBakbczyJMXfiPPCaotd1wtkEitu0V0o+P+UrvxzodDqdrkJ3tXdUGeuz/38AAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0002-4482-5122","institution":"TU Delft","correspondingAuthor":true,"prefix":"","firstName":"Julian","middleName":"","lastName":"Schumann","suffix":""},{"id":501304951,"identity":"4635665f-8986-4dd3-a541-08dcade02cda","order_by":1,"name":"Johan Engstrom","email":"","orcid":"","institution":"Waymo","correspondingAuthor":false,"prefix":"","firstName":"Johan","middleName":"","lastName":"Engstrom","suffix":""},{"id":501304952,"identity":"b48524be-089b-408f-be11-e7b9a2ee47cd","order_by":2,"name":"Leif Johnson","email":"","orcid":"","institution":"Waymo LLC","correspondingAuthor":false,"prefix":"","firstName":"Leif","middleName":"","lastName":"Johnson","suffix":""},{"id":501304953,"identity":"0ea44337-97d6-4560-af99-ea349dc3faff","order_by":3,"name":"Matthew O'Kelly","email":"","orcid":"","institution":"Waymo LLC","correspondingAuthor":false,"prefix":"","firstName":"Matthew","middleName":"","lastName":"O'Kelly","suffix":""},{"id":501304954,"identity":"2c17f741-b025-4c1e-a006-9ae5d65fc379","order_by":4,"name":"Joao Messias","email":"","orcid":"","institution":"Waymo LLC","correspondingAuthor":false,"prefix":"","firstName":"Joao","middleName":"","lastName":"Messias","suffix":""},{"id":501304955,"identity":"24a7b419-f980-494d-80dd-9c186430ad8f","order_by":5,"name":"Jens Kober","email":"","orcid":"https://orcid.org/0000-0001-7257-5434","institution":"Delft University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Jens","middleName":"","lastName":"Kober","suffix":""},{"id":501304956,"identity":"5e777a5c-2274-4e2c-8fd4-84b7ce2bb2ae","order_by":6,"name":"Arkady Zgonnikov","email":"","orcid":"https://orcid.org/0000-0002-6593-6948","institution":"Delft University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Arkady","middleName":"","lastName":"Zgonnikov","suffix":""}],"badges":[],"createdAt":"2025-06-02 18:50:26","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6804690/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6804690/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":89350181,"identity":"d272096a-d30e-487d-8e4c-5d47f851740f","added_by":"auto","created_at":"2025-08-19 05:57:57","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4708407,"visible":true,"origin":"","legend":"Article File","description":"","filename":"ActiveinferenceMainsubmission.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6804690/v1_covered_ba992988-fe6f-4620-864b-601351f8d5f4.pdf"},{"id":89349086,"identity":"9663f432-7b6d-476b-ac9f-b327ad0e1272","added_by":"auto","created_at":"2025-08-19 05:41:51","extension":"mp4","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":148000,"visible":true,"origin":"","legend":"Simulation underlying Figure 3A","description":"","filename":"Simulationfile3A.mp4","url":"https://assets-eu.researchsquare.com/files/rs-6804690/v1/306765ef21262cd9edc7245e.mp4"},{"id":89349084,"identity":"6e1cf2e6-eb02-4ba8-879b-bb0209ff78c2","added_by":"auto","created_at":"2025-08-19 05:41:51","extension":"mp4","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":184900,"visible":true,"origin":"","legend":"Simulation underlying Figure 3B","description":"","filename":"Simulationfile3B.mp4","url":"https://assets-eu.researchsquare.com/files/rs-6804690/v1/f97e2d55d1600cd58b14728d.mp4"},{"id":89349083,"identity":"3edabd87-4270-46ed-a6f3-dbee53c24c4b","added_by":"auto","created_at":"2025-08-19 05:41:51","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":80185,"visible":true,"origin":"","legend":"Description of supplementary files","description":"","filename":"ActiveInferenceSupplementarydescriptions.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6804690/v1/53062d8e44f7524fb13b2709.pdf"},{"id":89349089,"identity":"f5b82ad5-3d96-4f7e-ac73-5ec383e9295e","added_by":"auto","created_at":"2025-08-19 05:41:51","extension":"pdf","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":382047,"visible":true,"origin":"","legend":"\u003cp\u003eLaTeX Supplementary File\u003c/p\u003e","description":"","filename":"ActiveInferenceSupplementarymaterials.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6804690/v1/0c1f391ea9d3f93ad2898e24.pdf"},{"id":89349090,"identity":"73222618-8050-4d42-8671-e8da313355e9","added_by":"auto","created_at":"2025-08-19 05:41:51","extension":"mp4","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":330118,"visible":true,"origin":"","legend":"\u003cp\u003eSimulation underlying Figure 4A\u003c/p\u003e","description":"","filename":"Simulationfile4A.mp4","url":"https://assets-eu.researchsquare.com/files/rs-6804690/v1/90c603a1681727616f926269.mp4"},{"id":89349088,"identity":"d521377e-1b72-44ee-b9ff-a11423816e53","added_by":"auto","created_at":"2025-08-19 05:41:51","extension":"mp4","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":289701,"visible":true,"origin":"","legend":"\u003cp\u003eSimulation underlying Figure 4B\u003c/p\u003e","description":"","filename":"Simulationfile4B.mp4","url":"https://assets-eu.researchsquare.com/files/rs-6804690/v1/9251fa5e322988db593e5ce0.mp4"},{"id":89349380,"identity":"7faed212-fcc5-4abe-b5eb-52672a308c84","added_by":"auto","created_at":"2025-08-19 05:49:51","extension":"mp4","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":275242,"visible":true,"origin":"","legend":"\u003cp\u003eVideo referenced in the article\u003c/p\u003e","description":"","filename":"NoEAshallowincurisoncollision.mp4","url":"https://assets-eu.researchsquare.com/files/rs-6804690/v1/b7706f7cdf84c6f3c743c7a7.mp4"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Active inference as a unified model of collision avoidance behavior in human drivers","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-6804690/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6804690/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Collision avoidance -- involving a rapid threat detection and quick execution of the appropriate evasive maneuver -- is a critical aspect of driving. However, existing models of human collision avoidance behavior are fragmented, focusing on specific scenarios or only describing certain aspects of the avoidance behavior, such as response times.\r\nThis paper addresses these gaps by proposing a novel computational cognitive model of human collision avoidance behavior based on active inference.\r\nActive inference provides a unified approach to modeling human behavior: the minimization of free energy. Building on prior active inference work, our model incorporates established cognitive mechanisms such as evidence accumulation to simulate human responses in two distinct collision avoidance scenarios: front-to-rear lead vehicle braking and lateral incursion by an oncoming vehicle. We demonstrate that our model explains a wide range of previous empirical findings on human collision avoidance behavior. Specifically, the model closely reproduces both aggregate results from meta-analyses previously reported in the literature and detailed, scenario-specific effects observed in a recent driving simulator study, including response timing, maneuver selection, and execution. Our results highlight the potential of active inference as a unified framework for understanding and modeling human behavior in complex real-life driving tasks.","manuscriptTitle":"Active inference as a unified model of collision avoidance behavior in human drivers","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-19 05:41:46","doi":"10.21203/rs.3.rs-6804690/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"742c54ba-6a15-4e0d-9317-6f30bf5e6e70","owner":[],"postedDate":"August 19th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":53249345,"name":"Physical sciences/Engineering/Civil engineering"},{"id":53249346,"name":"Physical sciences/Engineering/Mechanical engineering"}],"tags":[],"updatedAt":"2026-05-06T15:17:35+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-19 05:41:46","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6804690","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6804690","identity":"rs-6804690","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}