Conversational agents : building trust, enhancing user experience, and fostering emotional connections

Abstract

The aim of this thesis is to explore the different feelings of users towards conversational agents through a questionnaire survey. The survey not only covers the various problems that can be caused by college students' over-reliance on text-based assistants (e.g. chatgpt), but also discusses how to optimise college students' perceptions of the user experience of text-based and voice-based conversation agents. This study is based on the results of exploring relevant literature and a questionnaire survey of students' attitudes towards text and voice assistants. Statistical tests and thematic analyses were used to provide an in-depth interpretation of the data in the questionnaire. My research reveals how conversational agents earn people's trust, however, over-trusting conversational agents may also raise the dilemma of people becoming too dependent. The survey data reveals that voice is more likely to win people's trust compared to text, and that voice conversational agent style has a significant effect on users' emotions and experiences. The limitations of this study focused on the age of the study and the data limitations of the questionnaire administered, due to the fact that a large amount of data support is required to gain insights into the participants' understandings and perspectives on conversational agents in order to obtain more trustworthy answers.