[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-273870-105":59,"doc-detail-273870-en":134},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":127,"head_meta":129,"extra_data":131,"updated_unix":133},105,"en","move-by-move-measuring-and-steering-how-llms-conduct-psychotherapy","Move by Move - Measuring and Steering How LLMs Conduct Psychotherapy","","Users increasingly rely on large language models for emotional support, yet the mechanisms behind their psychotherapy-like interactions remain underexplored. This work proposes a ten-move therapeutic ontology, using compact function-based categories grounded in MULTI-60, validated via an annotation campaign with five licensed psychologists and scaled with judge-based agreement. Applied to real counseling transcripts and model-led sessions, the study compares human clinicians versus frontier models, finding models overuse inquiry, omit psychoeducation, and follow human-initiated strategies without initiating them. Exposing the ontology as tools reduces deviation and improves turn-level alignment without fine-tuning.",{"@graph":69,"@context":126},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/move-by-move-measuring-and-steering-how-llms-conduct-psychotherapy/273870/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/move-by-move-measuring-and-steering-how-llms-conduct-psychotherapy/273870.png","ImageObject",300,407,{"name":92,"@type":93},"Ophelia","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-20","2026-09-15",true,{"@type":102,"interactionType":103,"userInteractionCount":44},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118,122],{"name":109,"@type":110,"acceptedAnswer":111},"What is the paper’s main goal regarding LLMs in psychotherapy?","Question",{"text":112,"@type":113},"To measure and understand how large language models conduct psychotherapy-like interactions, including which therapeutic interventions they favor or neglect and how their session conduct compares with trained clinicians.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"How is the ten-move therapeutic ontology developed and validated?",{"text":117,"@type":113},"The ontology uses compact, function-based categories grounded in the MULTI-60 inventory and is validated through an annotation campaign with five licensed psychologists, then scaled using judge-based agreement to match expert inter-annotator reliability.",{"name":119,"@type":110,"acceptedAnswer":120},"What differences are observed between LLM sessions and human clinician sessions?",{"text":121,"@type":113},"Models overuse inquiry by up to three times the human rate, neglect psychoeducation, and tend to be strongly anchored to context—carrying forward strategies started by a human clinician rather than initiating them themselves.",{"name":123,"@type":110,"acceptedAnswer":124},"How does exposing the ontology as tools affect model behavior?",{"text":125,"@type":113},"Exposing the ontology as tools roughly halves the deviation from the human move distribution and improves turn-level alignment with human therapists by about 7–9 percentage points, without any fine-tuning.","https://schema.org",{"og:url":83,"og:type":128,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":130,"canonical":83},"index,follow",{"doc_id":132,"site_id":62},273870,1789473085,{"code":4,"msg":5,"data":135},{"doc_id":132,"user_id":136,"nickname":92,"user_avatar":137,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":138,"file_id":139,"file_url":140,"file_type":141,"file_size":142,"view_count":44,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":143,"language":144,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":145,"faqs":146,"seo_title":147,"seo_description":67,"update_tm":133,"read_time":148,"preview_page_count":4},7971461741311,"https://ap-avatar.wpscdn.com/avatar/74000253aff267980c6?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779345379180704826","arXiv : 2608 .2 1325v 1 [cs .CL] 2 1 Aug 2026  \nMove by Move: Measuring and Steering How LLMs Conduct Psychotherapy  \nAfonso Baldo1, 3, *, Hugo Pitorro1, 3, *, Areti Vassilopoulos1, 2, Anabela C. Areias1, Maya D’Eon1 Fabíola Costa1, Ricardo Rei1, Nuno M. Guerreiro1  \n1Sword Health 2Yale University 3Instituto Superior Técnico  \n\\# [ai.research@sword.com](ai.research@sword.com)  \nAbstract  \nUsers increasingly turn to large language models for emotional support, yet little is known about how these models actually conduct a psychotherapy interaction. We introduce an ontology of ten therapeutic moves: compact, function-based categories grounded in the MULTI-60 inventory, validated through an annotation campaign with five licensed psychologists, and scaled with a judge-based approach that matches expert agreement. Applying it to real counseling transcripts and model-led sessions, we compare the move distributions between human clinicians and a panel of frontier models. Models over-use inquiry at up to three times the human rate, neglect psychoeducation, and are strongly context-anchored: they carry forward strategies initiated by a human clinician but rarely initiate them themselves. Exposing the ontology as a set of tools roughly halves the mean deviation from the human move distribution and improves turn-level alignment with human therapist by 7–9 percentage points, without any fine-tuning.  \n1 Introduction  \nMental-health disorders have risen sharply worldwide, with the global burden of anxiety and depressive disorders roughly doubling since the 1990s (GBD 2023 Mental Disorder Collaborators, 2026), while mental-health services and the clinical workforce have failed to keep pace with demand (van Heerdenet al., 2023) . Driven by lower cost, immediate availability, and privacy preferences, users increasingly turn to large language models (LLMs) for emotional support, counseling-adjacent conversation, and advice on interpersonal relationships (McCain et al., 2025; Rousmaniere et al., 2026) .  \nDespite known issues like sycophancy (Fanous et al., 2025), stemming form their training as harmless, helpful assistants (Ouyang et al., 2022; Bai et al., 2022), recent evaluations report LLM support as competitive to human responses, with controlled trials showing symptom reduction (Rollwage et al., 2026; Heinz et al., 2024) . Yet these outcome-level results leave a more basic question unanswered: we know little about how LLMs actually conduct a psychotherapy interaction, which interventions they favor, which they neglect, and how their conduct of a session compares with that of a trained clinician.  \nClinical psychology offers a natural lens for this question. Psychotherapy can be conceptualized as a sequential clinical decision-making process where the therapist continuously selects among alternative interventions as the conversation unfolds (Murphy, 2003; Goldfried, 1980) . Established coding frameworks, such as the Motivational Interviewing Skill Code (Miller et al., 2003a) and Hill’s Helping Skills (Hill, 2014a), operationalize this view by mapping each therapist utterance onto a discrete set of strategies, or skills (e.g., open questions, reflections, interpretations, challenges) . These frameworks underpin therapist training and supervision, letting experts systematically analyze which skills trainees deploy and provide precise, objective feedback (Hill et al., 2007, 2008; Rosengren, 2017) .  \nIn this work, we rely on clinical coding ontology both as a measurement instrument and as a steering mechanism. We introduce ten therapeutic moves: compact, function-based categories grounded in the MULTI-60 inventory (McCarthy and Barber, 2009) and validated through an annotation campaign with five licensed psychologists. Using an LLM judge that matches human inter-annotator agreement, we contrast the move distributions of human clinicians and a panel of models as therapists, both anchored in human session transcripts and leading them fully (Figu","cbCaimR2kkbGNWuc","https://ap.wps.com/l/cbCaimR2kkbGNWuc","pdf",941659,48,"English","# Abstract\n# Introduction\n## Motivation and problem framing\n## Therapy as sequential decision-making\n## Related coding frameworks\n# Method and contributions\n## Ten therapeutic moves ontology\n## LLM judge and move distribution analysis\n## Tool exposure and steering evaluation","[{\"question\":\"What is the paper’s main goal regarding LLMs in psychotherapy?\",\"answer\":\"To measure and understand how large language models conduct psychotherapy-like interactions, including which therapeutic interventions they favor or neglect and how their session conduct compares with trained clinicians.\"},{\"question\":\"How is the ten-move therapeutic ontology developed and validated?\",\"answer\":\"The ontology uses compact, function-based categories grounded in the MULTI-60 inventory and is validated through an annotation campaign with five licensed psychologists, then scaled using judge-based agreement to match expert inter-annotator reliability.\"},{\"question\":\"What differences are observed between LLM sessions and human clinician sessions?\",\"answer\":\"Models overuse inquiry by up to three times the human rate, neglect psychoeducation, and tend to be strongly anchored to context—carrying forward strategies started by a human clinician rather than initiating them themselves.\"},{\"question\":\"How does exposing the ontology as tools affect model behavior?\",\"answer\":\"Exposing the ontology as tools roughly halves the deviation from the human move distribution and improves turn-level alignment with human therapists by about 7–9 percentage points, without any fine-tuning.\"}]","Move by Move - Measuring and Steering How LLMs Conduct Psychotherapy | PDF",121]