Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Effectiveness of AI-based Cognitive Behavioural Therapy (AICBT) for Anxiety – A waiting list-controlled intervention study

Toiviainen, Toni LU (2025) PSYP01 20251
Department of Psychology
Abstract
Since the COVID-19 pandemic the need for mental healthcare has increased globally, which has been causing longer waiting times before treatment becomes a possibility. Many new innovations have been made to improve this, such as internet- and artificial intelligence-based treatments. The current study experimented the influence of TalkToAlba platform’s AI-cognitive behavioural therapy chatbot and its effects on patients with anxiety or matching symptoms. The main measures were anxiety, depression, and life-satisfaction, which were measured before, after, and 4-weeks after a 6-week long AICBT treatment, where participants took part in a minimum one AI-CBT meeting per week. This was compared to a control group, who instead of treatment had a... (More)
Since the COVID-19 pandemic the need for mental healthcare has increased globally, which has been causing longer waiting times before treatment becomes a possibility. Many new innovations have been made to improve this, such as internet- and artificial intelligence-based treatments. The current study experimented the influence of TalkToAlba platform’s AI-cognitive behavioural therapy chatbot and its effects on patients with anxiety or matching symptoms. The main measures were anxiety, depression, and life-satisfaction, which were measured before, after, and 4-weeks after a 6-week long AICBT treatment, where participants took part in a minimum one AI-CBT meeting per week. This was compared to a control group, who instead of treatment had a 6-week waiting-time before they got access to the AICBT chatbot. It was predicted that this treatment decreased anxiety and depression scores while increasing life-satisfaction. None of these predictions was confirmed despite changes occurring in the predicted direction. Exploratory analysis showed potential benefits of AICBT on anxiety symptoms, but only under correct conditions. Future research should focus on utilizing AICBT in collaboration with a clinic or clinician, as well as apply user-feedback for improved treatment efficacy. The study offers promising early results, despite them being nonsignificant, on AI-based mental health treatment, as well as feedback and future direction to improve said tools in the future.
Keywords: Artificial intelligence, AI-cognitive behavioural therapy, anxiety disorder, depression, waiting-list controlled, intervention study (Less)
Please use this url to cite or link to this publication:
author
Toiviainen, Toni LU
supervisor
organization
course
PSYP01 20251
year
type
H2 - Master's Degree (Two Years)
subject
keywords
Artificial intelligence, AI-cognitive behavioural therapy, anxiety disorder, depression, waiting-list controlled, intervention study
language
English
id
9212104
date added to LUP
2025-09-10 16:11:29
date last changed
2025-09-10 16:11:29
@misc{9212104,
  abstract     = {{Since the COVID-19 pandemic the need for mental healthcare has increased globally, which has been causing longer waiting times before treatment becomes a possibility. Many new innovations have been made to improve this, such as internet- and artificial intelligence-based treatments. The current study experimented the influence of TalkToAlba platform’s AI-cognitive behavioural therapy chatbot and its effects on patients with anxiety or matching symptoms. The main measures were anxiety, depression, and life-satisfaction, which were measured before, after, and 4-weeks after a 6-week long AICBT treatment, where participants took part in a minimum one AI-CBT meeting per week. This was compared to a control group, who instead of treatment had a 6-week waiting-time before they got access to the AICBT chatbot. It was predicted that this treatment decreased anxiety and depression scores while increasing life-satisfaction. None of these predictions was confirmed despite changes occurring in the predicted direction. Exploratory analysis showed potential benefits of AICBT on anxiety symptoms, but only under correct conditions. Future research should focus on utilizing AICBT in collaboration with a clinic or clinician, as well as apply user-feedback for improved treatment efficacy. The study offers promising early results, despite them being nonsignificant, on AI-based mental health treatment, as well as feedback and future direction to improve said tools in the future.
Keywords: Artificial intelligence, AI-cognitive behavioural therapy, anxiety disorder, depression, waiting-list controlled, intervention study}},
  author       = {{Toiviainen, Toni}},
  language     = {{eng}},
  note         = {{Student Paper}},
  title        = {{Effectiveness of AI-based Cognitive Behavioural Therapy (AICBT) for Anxiety – A waiting list-controlled intervention study}},
  year         = {{2025}},
}