PERCEIVED THERAPEUTIC BOND IN INTERACTIONS WITH MENTAL HEALTH CHATBOTS: A NARRATIVE REVIEW
Authors:
- Wojciech MarkiewiczWMWojciech Markiewicz
- Maria MarusińskaMMMaria Marusińska
- Michał KarpińskiMKMichał Karpiński
- Michał ParzniewskiMPMichał Parzniewski
- Wiktoria CzyżWCWiktoria Czyż
- Sabina KolawaSKSabina Kolawa
- Natalia OstruszkaNONatalia Ostruszka
- Arnold BorowiecABArnold Borowiec
- Julia PileckaJPJulia Pilecka
- JęDrzej WojciechowskiJWJęDrzej Wojciechowski
Open Access
- 2 July 2026
- journal article
- Published by RS Global Sp. z O.O. in International Journal of Innovative Technologies in Social Science
- Vol. 1 (3(51))
Abstract
Background: The rapid development of digital mental health interventions has led to the increasing use of AI-based chatbots as accessible alternatives or complements to traditional psychotherapy. While the therapeutic alliance is a well-established factor influencing treatment outcomes in conventional settings, its applicability to human–chatbot interactions remains unclear. Objective: This narrative review aims to examine existing evidence regarding the subjective experience of therapeutic bond in interactions with mental health chatbots and to explore its characteristics and potential clinical relevance. Methods: A narrative review of selected literature was conducted using databases such as PubMed and Google Scholar. Studies focusing on users’ subjective experiences and relational aspects of interaction with conversational agents were included. Results: Available evidence suggests that users can develop a form of therapeutic bond with mental health chatbots, characterized by perceived empathy, support, and emotional engagement. Both quantitative and qualitative studies indicate that users may attribute human-like qualities to chatbots and describe interactions in relational terms. However, the intensity and nature of this bond vary across individuals, with some users perceiving the interaction as purely instrumental. The bond appears to be limited in depth and may lack key features of traditional therapeutic relationships, such as reciprocity and genuine emotional responsiveness. Conclusions: Therapeutic bond in chatbot-based interventions may represent a distinct and simplified form of relational engagement shaped by the characteristics of human–AI interaction. Further research is needed to assess its long-term development and its relationship to clinical outcomes.Keywords
This publication has 12 references indexed in Scilit:
- The Digital Therapeutic Alliance With Mental Health Chatbots: Diary Study and Thematic AnalysisJMIR Mental Health, 2025
- Exploring relationship development with social chatbots: A mixed-method study of replikaComputers in Human Behavior, 2022
- The growing field of digital psychiatry: current evidence and the future of apps, social media, chatbots, and virtual realityWorld Psychiatry, 2021
- The alliance in adult psychotherapy: A meta-analytic synthesis.Psychotherapy, 2018
- An Empathy-Driven, Conversational Artificial Intelligence Agent (Wysa) for Digital Mental Well-Being: Real-World Data Evaluation Mixed-Methods StudyJMIR mHealth and uHealth, 2018
- Embodied Conversational Agents in Clinical Psychology: A Scoping ReviewJournal of Medical Internet Research, 2017
- Development and validation of a revised short version of the working alliance inventoryPsychotherapy Research, 2006
- Establishing and maintaining long-term human-computer relationshipsACM Transactions on Computer-Human Interaction, 2005
- The generalizability of the psychoanalytic concept of the working alliance.Psychotherapy, 1979
- ELIZA—a computer program for the study of natural language communication between man and machineCommunications of the ACM, 1966