Schema mode assessment through a conversational agent

Schema mode assessment through a conversational agent

Allaart, David (TU Delft Electrical Engineering, Mathematics and Computer Science)

Brinkman, W.P. (mentor)
Tielman, M.L. (graduation committee)
Lofi, C. (graduation committee)

Delft University of Technology

2021-03-17

Schema therapy is a psychotherapy for treatment of personality disorders and other psychological disorders. An important element of schema therapy is determining a patient's schema modes, a concept central to deciding the treatment approach. The objective of this work was to evaluate whether assessment of these schema modes is possible through a conversational agent. The agent designed held an interview style conversation, where first an open story was requested, which was then automatically analysed and followed up with evaluating questions. The results indicated the outcome of the agent was a significant predictor for a person's schema modes, in that a schema mode confirmed by the agent was 5.20 times more likely to be confirmed through standard means than a schema mode that was not confirmed by the agent. The results also hinted at a non-inferior user experience and time savings when using the agent instead of the questionnaire.

schema therapy
mental health
conversational agent
NLP
adaptive questionnaire

http://resolver.tudelft.nl/uuid:7388a711-2a75-40f6-a67a-c5e10a78bdaa

Student theses

master thesis

© 2021 David Allaart