The Impact of Vocal Communication and its Personalization on Intention to Use of Chatbots Using Behavioral Activation to Support Patients Experiencing Depression

Master thesis (2023)

Authors

K.Q. Doan Technology, Policy and Management

Contributors

N. Metoui (supervisor 1)
Aaron Yi Ding (supervisor 2)
G. Scholz Energy and Industry (supervisor 2)
Faculty
Technology, Policy and Management
Copyright: © 2023 Kevin Doan
Conversational agent Chatbot Trust Healthcare Personalization ChatGPT Quantitative analysis Adoption Survey design Design science Young adults Depression Mediation Mann-Whitney U test Interaction effect Speech Intention to use Voice Factorial design Between-subjects Affective trust Cognitive trust Ordinal logistic regression Behavioral activation Anthropomorphism
More Info
expand_more

Published Date

18-08-2023

Language

English

Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Faculty

Technology, Policy and Management

Abstract

The 21st century has seen a significant increase in the global prevalence of mental health problems, affecting almost a billion people. These conditions not only reduce the quality of life for individuals but also lead to stigmatization, discrimination, and social isolation. The COVID-19 pandemic has further exacerbated mental health issues, with depression and anxiety becoming the most common mental disorders among young workers. However, encouraging young adults to seek help for mental health is challenging.

To address this issue, the concept of digitalizing psychotherapies through mental health chatbot services is proposed as a potential solution. Giving the chatbot a voice to communicate is suggested to increase perceived trust and intention to use among users. A research question is posed: "How does vocal communication and its personalization in chatbots affect the intention to use of young adults with depression?"

The study conducted an online survey using Qualtrics with three groups of participants: the control group (text-based communication) and two treatment groups (vocal communication, and personalized vocal communication). The chatbot followed a rigid script based on a psychotherapy protocol called behavioral activation. Data analysis involved Mann-Whitney U tests, ordinal logistic regression, and one-way ANOVA for randomization check.

Findings from the study indicated that chatbots communicating through speech did not significantly influence trust compared to text-based chatbots. Personalization of the voice did not increase trust based on emotions and even negatively affected cognitive trust. Both cognitive and affective trust were positively associated with the intention to use chatbots. Cognitive trust indirectly mediated the relationship between personalized vocal communication and intention to use.

The study contributed to existing literature by exploring the adoption of conversational agents for mental health support, focusing on young adults with depression. It also provided insights into the use of speech in chatbots and the impact of personalization on trust and chatbot adoption. Additionally, the study shed light on the adoption of mental health chatbots in developing countries, which lack access to healthcare.

From a managerial and societal standpoint, the study's findings have implications for companies developing chatbots for mental healthcare. The insights gained can improve the design of mental health chatbots, potentially easing the burden on mental health professionals and contributing to long-term societal well-being. Moreover, the study highlights the potential benefits of chatbot accessibility for users in developing countries with limited access to healthcare.

Overall, the research provides valuable knowledge about the adoption of chatbots for mental health support, voice communication, personalization, and its implications for various demographics and regions. It also raises ethical considerations, particularly regarding gender-based biases when using gender-specific voices for chatbots.