{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,21]],"date-time":"2026-10-21T14:42:34Z","timestamp":1792593754855,"version":"4.5.0"},"reference-count":25,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR AI"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n Mentalization is a core human capacity involving interpretation of one\u2019s own and others\u2019 behavior in terms of underlying mental states. Within the mentalization-based treatment (MBT) framework, it is described along multiple dimensions, integrating cognitive, affective, relational, and regulatory processes. Large language models (LLMs) can generate linguistically reflective discourse, raising the question of whether the formal linguistic structure of mentalization can be reproduced without experiential and affective processes.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study aimed to assess whether an LLM can generate mentalization profiles structurally coherent with MBT dimensions and recognized as formally mentalizing by expert clinicians. Specifically, it evaluated the structural and argumentative coherence of model-generated profiles and the level of agreement among MBT-trained psychiatrists. It also examined the distinction between psychological mentalization and algorithmic reflectivity, conceptualized as a formally coherent but nonexperiential linguistic phenomenon.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n A comparative, descriptive methodological design was adopted. A total of 50 human-model dialogues were generated under standardized conditions. At the end of each interaction, the model produced a narrative mentalization profile structured along MBT dimensions. Five MBT-trained psychiatrists independently evaluated each profile using anonymized materials. Ratings covered 4 canonical MBT axes and 2 additional higher-order dimensions\u2014regulatory functions and synthetic profile. For each dimension, they assigned separate Likert-scale ratings for evaluative and argumentative coherence. Descriptive statistics were computed at dialogue level. Interrater reliability was assessed using intraclass correlation coefficients (ICCs), primarily ICC(3,k), with CIs. The association between evaluative and argumentative coherence was examined using Pearson correlation coefficients with 95% CIs, with Spearman rank correlation as a robustness check.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n \n Across dimensions, mean scores ranged from 3.63 (SD 0.63) to 3.98 (SD 0.31), indicating moderate-to-high structural coherence. Interrater reliability was moderate to good, with ICC(3,k) values ranging from 0.60 to 0.84 in the evaluative domain. The highest scores were observed for implicit-explicit and synthetic profile, whereas internal-external showed the lowest mean. Score distributions showed limited variability and concentration in the upper range of the scale. Evaluative and argumentative coherence were strongly associated across all MBT dimensions, with Pearson\n r<\/jats:italic>\n ranging from 0.826 to 0.967 (all\n P<\/jats:italic>\n <.001) and Spearman \u03c1 ranging from 0.866 to 0.991. Informal qualitative comments frequently described the outputs as linguistically organized and clinically interpretable, while also noting limited affective variability and contextual integration.\n <\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n The findings suggest that LLMs can reproduce the formal linguistic structure of reflective discourse, generating outputs perceived by clinicians as coherent and clinically interpretable across multiple dimensions of mentalization. However, within the present evaluative framework, this performance appears more consistent with structural language regularities than with evidence of experiential, affective, or intentional processes. These results support using LLMs as research and training tools for studying the linguistic aspects of mentalization, while highlighting the continuing importance of human clinical judgment in contexts requiring emotional, relational, and contextual integration.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/90750","type":"journal-article","created":{"date-parts":[[2026,9,1]],"date-time":"2026-09-01T14:39:52Z","timestamp":1788273592000},"page":"e90750-e90750","source":"Crossref","is-referenced-by-count":0,"title":["Algorithmic Reflectivity in Large Language Models: Methodological Study of Linguistic Structure in Mentalization"],"prefix":"10.2196","volume":"5","author":[{"ORCID":"https:\/\/orcid.org\/0009-0009-8935-9071","authenticated-orcid":false,"given":"Stefano","family":"Epifani","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-8135-4984","authenticated-orcid":false,"given":"Giuliano","family":"Castigliego","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-8574-8371","authenticated-orcid":false,"given":"Giuliano","family":"Razzicchia","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-4811-8725","authenticated-orcid":false,"given":"Elisabeth","family":"Seiwald-Sonderegger","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-8522-6098","authenticated-orcid":false,"given":"Laura","family":"Kecskem\u00e9ti","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,10,20]]},"reference":[{"key":"R1","unstructured":"Fonagy P Gergely G Jurist EL Target M . Affect Regulation, Mentalization, and the Development of the Self. Other Press; 2002. [doi: 978-0429910647]"},{"key":"R2","doi-asserted-by":"crossref","unstructured":"Allen JG Fonagy P Bateman AW . Mentalizing in Clinical Practice. American Psychiatric Publishing; 2008. [doi: 10.1176\/appi.books.9781615370887]","DOI":"10.1176\/appi.books.9781615370887"},{"key":"R3","doi-asserted-by":"crossref","unstructured":"Bateman AW Fonagy P . Mentalization-Based Treatment for Personality Disorders: A Practical Guide. 2nd ed. Oxford University Press; 2016. [doi: 10.1093\/med:psych\/9780199680375.001.0001]","DOI":"10.1093\/med:psych\/9780199680375.001.0001"},{"issue":"6","key":"R4","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.2218523120","article-title":"Using cognitive psychology to understand GPT-3","volume":"120","author":"Binz","journal-title":"Proc Natl Acad Sci U S A"},{"issue":"6","key":"R5","doi-asserted-by":"publisher","first-page":"517","DOI":"10.1016\/j.tics.2024.01.011","article-title":"Dissociating language and thought in large language models","volume":"28","author":"Mahowald","journal-title":"Trends Cogn Sci (Regul Ed)"},{"key":"R6","unstructured":"Kosinski M . Evaluating large language models in theory of mind tasks. arXiv. Preprint posted online on Feb 4, 2023. URL: https:\/\/arxiv.org\/abs\/2302.02083 [Accessed 28-09-2026]"},{"issue":"7","key":"R7","doi-asserted-by":"publisher","first-page":"1285","DOI":"10.1038\/s41562-024-01882-z","article-title":"Testing theory of mind in large language models and humans","volume":"8","author":"Strachan","journal-title":"Nat Hum Behav"},{"key":"R8","unstructured":"Ullman TD . Large language models fail on trivial alterations to theory-of-mind tasks. arXiv. Preprint posted online on Feb 16, 2023. [doi: 10.48550\/arXiv.2302.08399]"},{"key":"R9","doi-asserted-by":"publisher","DOI":"10.3389\/fpsyt.2023.1234397","article-title":"The plasticity of ChatGPT\u2019s mentalizing abilities: personalization for personality structures","volume":"14","author":"Hadar-Shoval","journal-title":"Front Psychiatry"},{"issue":"4","key":"R10","doi-asserted-by":"publisher","first-page":"531","DOI":"10.1016\/j.neuron.2006.05.001","article-title":"The neural basis of mentalizing","volume":"50","author":"Frith","journal-title":"Neuron"},{"issue":"5","key":"R11","doi-asserted-by":"publisher","first-page":"675","DOI":"10.1038\/nn.3085","article-title":"The neuroscience of empathy: progress, pitfalls and promise","volume":"15","author":"Zaki","journal-title":"Nat Neurosci"},{"issue":"11","key":"R12","doi-asserted-by":"publisher","first-page":"512","DOI":"10.1016\/j.tics.2011.09.003","article-title":"What is so special about embodied simulation?","volume":"15","author":"Gallese","journal-title":"Trends Cogn Sci (Regul Ed)"},{"key":"R13","doi-asserted-by":"publisher","DOI":"10.2196\/79156","article-title":"Mentalizing without a mind: psychotherapeutic potential of generative AI","volume":"27","author":"Yirmiya","journal-title":"J Med Internet Res"},{"key":"R14","unstructured":"Radford A Jozefowicz R Sutskever I . Learning to generate reviews and discovering sentiment. arXiv. Preprint posted online on Apr 5, 2017. [doi: 10.48550\/arXiv.1704.01444]"},{"issue":"2","key":"R15","doi-asserted-by":"publisher","first-page":"15","DOI":"10.1109\/MIS.2013.30","article-title":"New avenues in opinion mining and sentiment analysis","volume":"28","author":"Cambria","journal-title":"IEEE Intell Syst"},{"key":"R16","doi-asserted-by":"crossref","unstructured":"Devlin J Chang MW Lee K Toutanova K . BERT: pre-training of deep bidirectional transformers for language understanding. Presented at: Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics; Jun 2-7, 2019:4171-4186; Minneapolis, Minnesota. 2019.[doi: 10.18653\/v1\/N19-1423]","DOI":"10.18653\/v1\/N19-1423"},{"key":"R17","author":"Brown","journal-title":"Adv Neural Inf Process Syst"},{"key":"R18","unstructured":"Liang P Bommasani R Lee T et al. Holistic evaluation of language models. arXiv. Preprint posted online on Nov 16, 2022. [doi: 10.48550\/arXiv.2211.09110]"},{"issue":"2","key":"R19","doi-asserted-by":"publisher","first-page":"420","DOI":"10.1037\/\/0033-2909.86.2.420","article-title":"Intraclass correlations: uses in assessing rater reliability","volume":"86","author":"Shrout","journal-title":"Psychol Bull"},{"issue":"2","key":"R20","doi-asserted-by":"publisher","first-page":"155","DOI":"10.1016\/j.jcm.2016.02.012","article-title":"A guideline of selecting and reporting intraclass correlation coefficients for reliability research","volume":"15","author":"Koo","journal-title":"J Chiropr Med"},{"issue":"9","key":"R21","doi-asserted-by":"publisher","first-page":"1331","DOI":"10.1002\/sim.1108","article-title":"Sample size requirements for estimating intraclass correlations with desired precision","volume":"21","author":"Bonett","journal-title":"Stat Med"},{"key":"R22","unstructured":"Gurnee W Tegmark M . Language models represent space and time. arXiv. Preprint posted online on Oct 3, 2023. [doi: 10.48550\/arXiv.2310.02207]"},{"issue":"3","key":"R23","doi-asserted-by":"publisher","first-page":"372","DOI":"10.1037\/a0036505","article-title":"The role of mentalizing and epistemic trust in the therapeutic relationship","volume":"51","author":"Fonagy","journal-title":"Psychotherapy (Chic)"},{"key":"R24","doi-asserted-by":"publisher","DOI":"10.2196\/87367","article-title":"Shoggoths, sycophancy, psychosis, oh my: rethinking large language model use and safety","volume":"27","author":"Clegg","journal-title":"J Med Internet Res"},{"key":"R25","doi-asserted-by":"crossref","unstructured":"Malmqvist L . Sycophancy in large language models: causes and mitigations. In: Arai K , editor. Intelligent Computing: Proceedings of the 2025 Computing Conference, Volume 4. Springer Cham; 2025:61-74. [doi: 10.1007\/978-3-031-92611-2_5]","DOI":"10.1007\/978-3-031-92611-2_5"}],"container-title":["JMIR AI"],"original-title":[],"language":"en","deposited":{"date-parts":[[2026,10,21]],"date-time":"2026-10-21T14:11:18Z","timestamp":1792591878000},"score":1,"resource":{"primary":{"URL":"https:\/\/ai.jmir.org\/2026\/1\/e90750"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10,20]]},"references-count":25,"URL":"https:\/\/doi.org\/10.2196\/90750","relation":{},"ISSN":["2817-1705"],"issn-type":[{"value":"2817-1705","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,20]]},"article-number":"v5i6e90750"}}