My research aims to establish a transformative foundation for designing human-machine social interaction that centers on multi-faceted human experience, resonates with human values, and foster human aspirations. I study machines and humans as coupled systems rather than separate entities, and perceive machines as socially constructed entities beyond mere algorithms, code, and data. As highly social and technologically sophisticated interfaces, personified social agents powered by artificial intelligence (AI) such as social robots are the central focus of my work.
Research Themes
My research examines the following four research themes via a holistic, iterative process.
Conceptual Formulation: How can we envision a conceptual paradigm shift in social human-agent interactions from the current predominant design?
I develop conceptual tools and explore conceptual grounds for next-gen social human-agent interaction by integrating insights from yet-to-be-tapped theories from social sciences, such as flow theory and human mimicry.
Innovative Interaction: How can we create agent-augmented experiences to enhance human functions and human-human communication?
I design equitable and inclusive human-agent interaction innovation in-the-wild, such as agents as social catalysts to enhance human-human interaction and agents as reciprocal peers that can switch their social roles.
Scientific Insights: How do agents shape human cognition, emotions, and behaviors within and after interactions?
I identify the cross-disciplinary factors in interactions that critically mediate or moderate these effects and build psychological mechanisms underlying human-agent interactions. The facets of the agent’s impact I explore include human learning and performance, conversational dynamics, behavior-based emotions, and preference for technology.
Technological Enablers: What advanced social sophistication should agents acquire to support more value-centric interactions for human aspirations?
I develop computational methods building upon deep learning and reinforcement learning to enable agents to understand affective and social dynamics in human-human interactions and tailor its behaviors to individuals and groups. I also collect real-world interaction datasets and extract insights useful for model design and trainings.
Research Areas
Human-computer Interaction; Human-centered AI; Social Robotics; Affective Computing; Cognitive Science; Social Signal Processing; Psychology of Technology; AI Ethics;
Research Approach & Methods
My approach spans a broad spectrum of technical and methodological tools, from design methodologies to randomized control trials, and from machine learning techniques to social robotic interfaces. I design cutting-edge interactive technologies, such as social robots, and deploy them in K-12 public schools and family homes for months. Through the application of varied methods, the integration of interdisciplinary viewpoints such as psychology, computing, and design, and the involvement of diverse stakeholders, including users and developers, my research employs cross-disciplinary lenses to foster diversified synergistic insights.
Research Skills
Human Subjects Research; Qualitative analysis; Applied Machine Learning; Statistics in Social Science; Human-centered Design; Randomized Controlled Trial; Complex Network Analysis; Crowdsourcing; Oral History