Human-AI-Interaction | Conversational Agents - Users’ Choice of Interaction Modality
AI-based user assistance systems such as conversational agents (CAs) increasingly permeate our lives and offer us assistance with a myriad of tasks. By assisting professionals in decision-making or even taking over complete tasks from them, CAs assume an important role in the work context. Similarly, in private life, CAs such as Apple’s Siri or Amazon’s Alexa become personal assistants and make users’ lives easier, for example by providing the latest weather forecast or playing users’ favorite music. Interaction with CAs can be text- or voice-based. Yet CA use remains below the expected level and user acceptance is limited, implying enormous unrealized potential.
Therefore, a deeper understanding of the conditions under which voice- or text-based CAs are used will be important to identify ways to foster their use and inform user-centric CA design.
The thesis aims at pursuing an empirical approach to understand which factors influence the choice of interaction mode (i.e., text or voice) when interacting with a CA. In particular, situational factors and user characteristics that affect users’ choice of interaction mode shall be explored using quantitative research methods such as a survey or experiment. You will develop a research model and conduct a survey/experiment to test your hypotheses. The collected data will be analyzed using regression analyses, hypothesis testing, or similar statistical methods.
- Interest in human-computer interaction and conversational agents
- High motivation to work on interesting real-world problems
- First experience in data analysis with R or Python
- Good English skills as the thesis will be written in English
- Self-organized and goal-oriented working mode, including the motivation to bring own ideas
We offer you a challenging research topic, close supervision, and the opportunity to develop practical and theoretical skills. If you are interested, please send an email to Lara Riefle along with a transcript of records and CV.