Exploiting Large Language Models for Diagnosing Autism Associated Language Disorders and Identifying Distinct Features

Exploiting Large Language Models for Diagnosing Autism Associated Language Disorders and Identifying Distinct Features | 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 Exploiting Large Language Models for Diagnosing Autism Associated Language Disorders and Identifying Distinct Features Chuanbo Hu, Wenqi Li, Mindi Ruan, Xiangxu Yu, Shalaka Deshpande, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6931837/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 16 Dec, 2025 Read the published version in npj Digital Medicine → Version 1 posted 6 You are reading this latest preprint version Abstract Diagnosing language disorders associated with autism is a complex challenge, often hampered by the subjective nature and variability of traditional assessment methods. In this study, we explored Large Language Models (LLMs) to overcome the speed and precision obstacles by enhancing sensitivity and profiling linguistic features for autism diagnosis. This research utilizes natural language understanding capabilities of LLMs to simplify and improve the diagnostic process, focusing on identifying autism-related language patterns. We showed that the proposed method demonstrated improvements over the baseline models, with over a 10% increase in both sensitivity and positive predictive value in a zero-shot learning configuration. Combining accuracy and applicability, the framework could serve as a valuable supplementary tool within the diagnostic process for ASD-related language patterns. We identified ten key features of autism-associated language disorders across scenarios. Features such as echolalia, pronoun reversal, and atypical language usage play a critical role in diagnosing ASD and informing tailored treatment plans. Biological sciences/Neuroscience Health sciences/Biomarkers Physical sciences/Mathematics and computing Autism spectrum disorder (ASD) Language deficits Identified linguistic features Large language models (LLMs) Generative Pre-trained Transformer (GPT) Full Text Additional Declarations No competing interests reported. Supplementary Files AuthorContributionStatement.docx Dataavailabilitystatement.docx Fundingacknowledgement.docx Cite Share Download PDF Status: Published Journal Publication published 16 Dec, 2025 Read the published version in npj Digital Medicine → Version 1 posted Editorial decision: Revision requested 22 Aug, 2025 Reviewers agreed at journal 30 Jul, 2025 Reviewers invited by journal 27 Jul, 2025 Editor assigned by journal 18 Jul, 2025 Submission checks completed at journal 17 Jul, 2025 First submitted to journal 19 Jun, 2025 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-6931837","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":491603156,"identity":"80cd9953-603f-4b80-a151-8429118ab7f1","order_by":0,"name":"Chuanbo Hu","email":"","orcid":"","institution":"University at Albany, State University of New York","correspondingAuthor":false,"prefix":"","firstName":"Chuanbo","middleName":"","lastName":"Hu","suffix":""},{"id":491603157,"identity":"32bc8b91-2fd5-4d24-95e8-010d2517a634","order_by":1,"name":"Wenqi Li","email":"","orcid":"","institution":"University at Albany, State University of New York","correspondingAuthor":false,"prefix":"","firstName":"Wenqi","middleName":"","lastName":"Li","suffix":""},{"id":491603158,"identity":"1b9f7ef7-3f44-480f-850f-73a7940dcc56","order_by":2,"name":"Mindi Ruan","email":"","orcid":"","institution":"West Virginia University","correspondingAuthor":false,"prefix":"","firstName":"Mindi","middleName":"","lastName":"Ruan","suffix":""},{"id":491603159,"identity":"6b4eac98-4241-4cb9-a18e-d7ca0bc1a30f","order_by":3,"name":"Xiangxu Yu","email":"","orcid":"","institution":"Washington University in St. Louis","correspondingAuthor":false,"prefix":"","firstName":"Xiangxu","middleName":"","lastName":"Yu","suffix":""},{"id":491603160,"identity":"65f65e11-2e14-42b0-ba10-6ba8cf9421d9","order_by":4,"name":"Shalaka Deshpande","email":"","orcid":"","institution":"Camas High School","correspondingAuthor":false,"prefix":"","firstName":"Shalaka","middleName":"","lastName":"Deshpande","suffix":""},{"id":491603161,"identity":"a767309f-160e-433c-8a40-4ea58417db71","order_by":5,"name":"Lynn K. Paul","email":"","orcid":"","institution":"California Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Lynn","middleName":"K.","lastName":"Paul","suffix":""},{"id":491603162,"identity":"fde3fa8f-de6c-4c71-b06f-777260495c2c","order_by":6,"name":"Shuo Wang","email":"","orcid":"","institution":"Washington University in St. Louis","correspondingAuthor":false,"prefix":"","firstName":"Shuo","middleName":"","lastName":"Wang","suffix":""},{"id":491603163,"identity":"0f36ae09-a3fa-48ec-b07f-73efcaea2986","order_by":7,"name":"Xin Li","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAzklEQVRIiWNgGAWjYDACdsYGgw8VDDIGEC4zEVqYGRsKZ5xh4CFFCwPDZ942UrQYHGZu3Dhznh2POfvZYxIMFdaJDYS0SDYzNht83JbMY9mTlybBcCadsBZ+ZsY2w5nbmHkMDuSYSTC2HSashY2Zsf0375x6HoPzb4Ba/hGhBWhLgzFvw2EegxsgWxqI0AL0S4PhjGPHeSxnvDG2SDiWbkxQi8Hx9gcGH2qq5cz5cwxvfKixliWoBRmwSCSQohwEmD+QqmMUjIJRMApGBgAASYw6qxixfOwAAAAASUVORK5CYII=","orcid":"","institution":"University at Albany, State University of New York","correspondingAuthor":true,"prefix":"","firstName":"Xin","middleName":"","lastName":"Li","suffix":""}],"badges":[],"createdAt":"2025-06-19 13:53:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6931837/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6931837/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41746-025-02133-9","type":"published","date":"2025-12-16T15:57:46+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":98813949,"identity":"1f2adb31-d58b-4c0a-b89f-572f683c312e","added_by":"auto","created_at":"2025-12-22 16:08:20","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3684703,"visible":true,"origin":"","legend":"","description":"","filename":"ChatGPTAutismlanguagedisorderNPJ.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6931837/v1_covered_f8ba4d3d-0ff7-42b1-8cad-37796d50ef45.pdf"},{"id":87834527,"identity":"3d937adc-671a-4aa1-bf85-06daa09a70a2","added_by":"auto","created_at":"2025-07-29 13:01:49","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":13447,"visible":true,"origin":"","legend":"","description":"","filename":"AuthorContributionStatement.docx","url":"https://assets-eu.researchsquare.com/files/rs-6931837/v1/01257d438932032eadee2724.docx"},{"id":87834525,"identity":"0637963d-fbe1-4ad3-85ce-6183e8eda44d","added_by":"auto","created_at":"2025-07-29 13:01:49","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":14416,"visible":true,"origin":"","legend":"","description":"","filename":"Dataavailabilitystatement.docx","url":"https://assets-eu.researchsquare.com/files/rs-6931837/v1/1d33343ab76a7729d1d61b7a.docx"},{"id":87835284,"identity":"2f6d0861-4b22-4e27-bba7-acfcdcc3feee","added_by":"auto","created_at":"2025-07-29 13:09:49","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":13605,"visible":true,"origin":"","legend":"","description":"","filename":"Fundingacknowledgement.docx","url":"https://assets-eu.researchsquare.com/files/rs-6931837/v1/30a85e762ecc0819e3644fde.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Exploiting Large Language Models for Diagnosing Autism Associated Language Disorders and Identifying Distinct Features","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"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":"npj-digital-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"npjdigitalmed","sideBox":"Learn more about [npj Digital Medicine](http://www.nature.com/npjdigitalmed/)","snPcode":"41746","submissionUrl":"https://submission.springernature.com/new-submission/41746/3","title":"npj Digital Medicine","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Autism spectrum disorder (ASD), Language deficits, Identified linguistic features, Large language models (LLMs), Generative Pre-trained Transformer (GPT)","lastPublishedDoi":"10.21203/rs.3.rs-6931837/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6931837/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eDiagnosing language disorders associated with autism is a complex challenge, often hampered by the subjective nature and variability of traditional assessment methods. In this study, we explored Large Language Models (LLMs) to overcome the speed and precision obstacles by enhancing sensitivity and profiling linguistic features for autism diagnosis. This research utilizes natural language understanding capabilities of LLMs to simplify and improve the diagnostic process, focusing on identifying autism-related language patterns. We showed that the proposed method demonstrated improvements over the baseline models, with over a 10% increase in both sensitivity and positive predictive value in a zero-shot learning configuration. Combining accuracy and applicability, the framework could serve as a valuable supplementary tool within the diagnostic process for ASD-related language patterns. We identified ten key features of autism-associated language disorders across scenarios. Features such as echolalia, pronoun reversal, and atypical language usage play a critical role in diagnosing ASD and informing tailored treatment plans.\u003c/p\u003e","manuscriptTitle":"Exploiting Large Language Models for Diagnosing Autism Associated Language Disorders and Identifying Distinct Features","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-29 13:01:44","doi":"10.21203/rs.3.rs-6931837/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-08-22T17:17:54+00:00","index":"","fulltext":""},{"type":"reviewerAgreed","content":"236953693492577257136108422245655240251","date":"2025-07-30T22:35:13+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-07-27T23:40:59+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-07-18T18:20:14+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-07-17T16:04:00+00:00","index":"","fulltext":""},{"type":"submitted","content":"npj Digital Medicine","date":"2025-06-19T13:42:17+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"npj-digital-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"npjdigitalmed","sideBox":"Learn more about [npj Digital Medicine](http://www.nature.com/npjdigitalmed/)","snPcode":"41746","submissionUrl":"https://submission.springernature.com/new-submission/41746/3","title":"npj Digital Medicine","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"6374091d-4b2f-4979-9217-7cb3277ea056","owner":[],"postedDate":"July 29th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":52191410,"name":"Biological sciences/Neuroscience"},{"id":52191411,"name":"Health sciences/Biomarkers"},{"id":52191412,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2025-12-22T16:01:49+00:00","versionOfRecord":{"articleIdentity":"rs-6931837","link":"https://doi.org/10.1038/s41746-025-02133-9","journal":{"identity":"npj-digital-medicine","isVorOnly":false,"title":"npj Digital Medicine"},"publishedOn":"2025-12-16 15:57:46","publishedOnDateReadable":"December 16th, 2025"},"versionCreatedAt":"2025-07-29 13:01:44","video":"","vorDoi":"10.1038/s41746-025-02133-9","vorDoiUrl":"https://doi.org/10.1038/s41746-025-02133-9","workflowStages":[]},"version":"v1","identity":"rs-6931837","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6931837","identity":"rs-6931837","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}