Teaching-Learning Interaction Design

2021 - present

Different forms of intelligent systems learn about their users to provide more personalized services through the using process. However, the learning processes of these systems are designed with little consideration of user agency, resulting in hindering users from making the best use of the system. In this project, we propose Teaching-Learning Interaction (TLI) as a new form of interaction that supports user agency by letting users reflectively shepherd an intelligent system’s manner of learning. To solidify TLI, we investigate user needs and experiences in teaching their intelligent systems from various domains (i.e., recommender systems, autonomous vehicles, etc.).

For example, while users have little control of modifying the driving styles of current autonomous vehicles (AVs), the concept of TLI enables users to guide their driving-style preferences to adjust AV’s driving more fit to their needs. In our proposal, user agency is a core driving factor that is in control of AV’s learning process. While utilizing the empirical findings to develop and evaluate more detailed interaction scenarios and design guidelines for the intelligent systems, we aim to collect multiple design cases adequate for TLI proposal. Accompanied by TLI and user agency, users will be able to better personalize services for themselves.