Accompanying Reflection Processes by an AI-Based StudiCoachBot: A Study on Rapport Building in Human-Machine Coaching Using Self Disclosure

Abstract

The establishment of a working alliance, which in human-machine interaction is comparable to the establishment of rapport, is considered a central factor for successful coaching. The goal of this study is to gain insights into how such a relationship can be established between a coaching chatbot and students. One factor considered to establish rapport is the concept of self disclosure. To this end, the self disclosure of a coaching chatbot is examined as a factor influencing student self disclosure and perceived rapport. An interaction script for coaching via chatbot on the topic of exam anxiety was developed and tested with students of the TH Köln/University of Applied Sciences in a Wizard-of-Oz experiment. Following the coaching, a survey and an interview took place. The results show that students disclosed themselves to the chatbot and built a relationship with it. However, contrary to the assumption, more self disclosure and rapport was found on average in the control group (information disclosure by the chatbot) than in the experimental group (self disclosure by the chatbot). This suggests that general information disclosure seems to be sufficient and tends to generate even more self disclosure and rapport among students than chatbot self disclosure. The results also suggest that chatbot coaching in its current form is already well received by students and helps them reflect on their exam anxiety. By having the coaching done by a machine, students find it easy to open up. There is a great willingness to use it.