The effect of ChatGPT intervention on depressive and anxious states mediated by a decrease in perceived invalidation and the frequency of automatic thoughts – A pilot study
Abstract
The purpose of this study is to confirm whether a conversation with the LLM agent ChatGPT is capable of reducing depressive and anxious states in the general healthy adult population of Romania and which underlying mechanism (mediator), either reduction in perceived emotional invalidation or frequency of automatic thoughts, is responsible for the change in depressive and anxious states. It is expected that a statistically significant reduction in depressive and anxious states will occur in the experimental condition, in which participants will have to converse in a total of eleven prompts with ChatGPT about personal subjects, compared with the control condition, where participants converse in a total of eleven prompts with ChatGPT about non-personal subjects. It is also expected that in the experimental condition there will be a statistically significant reduction in depressive and anxious states from baseline measurement to post-intervention measurement.
Fields of Study & Subjects
Digital Mental Health Interventions
Anxiety
Depression
Psychometrics
Treatment
Cognitive Processes
Social Robot Interaction and HRI
Publication details
- Publication date
- 1 January 2026
- Language
- EN
- Document type
- Preprint