COMPASS: Computational mapping of patient-therapist alliance strategies with language modeling

Abstract The therapeutic working alliance is a critical predictor of psychotherapy success. Traditionally, working alliance assessment relies on questionnaires completed by both therapists and patients. In this paper, we present COMPASS, a novel framework to directly infer the therapeutic working al...

Full description

Saved in:
Bibliographic Details
Main Authors: Baihan Lin, Djallel Bouneffouf, Yulia Landa, Rachel Jespersen, Cheryl Corcoran, Guillermo Cecchi
Format: Article
Language:English
Published: Nature Publishing Group 2025-05-01
Series:Translational Psychiatry
Online Access:https://doi.org/10.1038/s41398-025-03379-3
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849310292827176960
author Baihan Lin
Djallel Bouneffouf
Yulia Landa
Rachel Jespersen
Cheryl Corcoran
Guillermo Cecchi
author_facet Baihan Lin
Djallel Bouneffouf
Yulia Landa
Rachel Jespersen
Cheryl Corcoran
Guillermo Cecchi
author_sort Baihan Lin
collection DOAJ
description Abstract The therapeutic working alliance is a critical predictor of psychotherapy success. Traditionally, working alliance assessment relies on questionnaires completed by both therapists and patients. In this paper, we present COMPASS, a novel framework to directly infer the therapeutic working alliance from the natural language used in psychotherapy sessions. Our approach leverages advanced large language models (LLMs) to analyze session transcripts and map them to distributed representations. These representations capture the semantic similarities between the dialogues and psychometric instruments, such as the Working Alliance Inventory. Analyzing a dataset of over 950 sessions spanning diverse psychiatric conditions -- including anxiety (N = 498), depression (N = 377), schizophrenia (N = 71), and suicidal tendencies (N = 12) -- collected between 1970 and 2012, we demonstrate the effectiveness of our method in providing fine-grained mapping of patient-therapist alignment trajectories, offering interpretable insights for clinical practice, and identifying emerging patterns related to the condition being treated. By employing various deep learning-based topic modeling techniques in combination with prompting generative language models, we analyze the topical characteristics of different psychiatric conditions and how these topics evolve during each turn of the conversation. This integrated framework enhances the understanding of therapeutic interactions, enables timely feedback for therapists on the quality of therapeutic relationships, and provides clear, actionable insights to improve the effectiveness of psychotherapy.
format Article
id doaj-art-aada23d170634e2eb9bb65bfd60f4e8b
institution Kabale University
issn 2158-3188
language English
publishDate 2025-05-01
publisher Nature Publishing Group
record_format Article
series Translational Psychiatry
spelling doaj-art-aada23d170634e2eb9bb65bfd60f4e8b2025-08-20T03:53:46ZengNature Publishing GroupTranslational Psychiatry2158-31882025-05-0115111510.1038/s41398-025-03379-3COMPASS: Computational mapping of patient-therapist alliance strategies with language modelingBaihan Lin0Djallel Bouneffouf1Yulia Landa2Rachel Jespersen3Cheryl Corcoran4Guillermo Cecchi5Department of Artificial Intelligence and Human Health, Icahn School of Medicine at Mount SinaiIBM Research, T.J. Watson Research CenterDepartment of Psychiatry, Icahn School of Medicine at Mount SinaiDepartment of Psychiatry, Icahn School of Medicine at Mount SinaiDepartment of Psychiatry, Icahn School of Medicine at Mount SinaiIBM Research, T.J. Watson Research CenterAbstract The therapeutic working alliance is a critical predictor of psychotherapy success. Traditionally, working alliance assessment relies on questionnaires completed by both therapists and patients. In this paper, we present COMPASS, a novel framework to directly infer the therapeutic working alliance from the natural language used in psychotherapy sessions. Our approach leverages advanced large language models (LLMs) to analyze session transcripts and map them to distributed representations. These representations capture the semantic similarities between the dialogues and psychometric instruments, such as the Working Alliance Inventory. Analyzing a dataset of over 950 sessions spanning diverse psychiatric conditions -- including anxiety (N = 498), depression (N = 377), schizophrenia (N = 71), and suicidal tendencies (N = 12) -- collected between 1970 and 2012, we demonstrate the effectiveness of our method in providing fine-grained mapping of patient-therapist alignment trajectories, offering interpretable insights for clinical practice, and identifying emerging patterns related to the condition being treated. By employing various deep learning-based topic modeling techniques in combination with prompting generative language models, we analyze the topical characteristics of different psychiatric conditions and how these topics evolve during each turn of the conversation. This integrated framework enhances the understanding of therapeutic interactions, enables timely feedback for therapists on the quality of therapeutic relationships, and provides clear, actionable insights to improve the effectiveness of psychotherapy.https://doi.org/10.1038/s41398-025-03379-3
spellingShingle Baihan Lin
Djallel Bouneffouf
Yulia Landa
Rachel Jespersen
Cheryl Corcoran
Guillermo Cecchi
COMPASS: Computational mapping of patient-therapist alliance strategies with language modeling
Translational Psychiatry
title COMPASS: Computational mapping of patient-therapist alliance strategies with language modeling
title_full COMPASS: Computational mapping of patient-therapist alliance strategies with language modeling
title_fullStr COMPASS: Computational mapping of patient-therapist alliance strategies with language modeling
title_full_unstemmed COMPASS: Computational mapping of patient-therapist alliance strategies with language modeling
title_short COMPASS: Computational mapping of patient-therapist alliance strategies with language modeling
title_sort compass computational mapping of patient therapist alliance strategies with language modeling
url https://doi.org/10.1038/s41398-025-03379-3
work_keys_str_mv AT baihanlin compasscomputationalmappingofpatienttherapistalliancestrategieswithlanguagemodeling
AT djallelbouneffouf compasscomputationalmappingofpatienttherapistalliancestrategieswithlanguagemodeling
AT yulialanda compasscomputationalmappingofpatienttherapistalliancestrategieswithlanguagemodeling
AT racheljespersen compasscomputationalmappingofpatienttherapistalliancestrategieswithlanguagemodeling
AT cherylcorcoran compasscomputationalmappingofpatienttherapistalliancestrategieswithlanguagemodeling
AT guillermocecchi compasscomputationalmappingofpatienttherapistalliancestrategieswithlanguagemodeling