NTU Embodied Dynamics Lab
Embodied human–AI interaction
We study how people understand, communicate with, and form relationships with intelligent machines. The lab is based at the Wee Kim Wee School of Communication and Information, Nanyang Technological University.
How do people interpret intelligent machines?
People form expectations about an intelligent system before an exchange begins. Its appearance, movement, voice, distance, and interface all affect whether it is understood as a tool, an agent, or a social partner. We study how these cues shape trust, agency, affect, and the meaning people assign to an interaction.
It begins as a word: kinematic.
A body is a set of relationships.
It takes up space.
And it moves. It greets you.
How do embodiment and repeated interaction shape relationships?
A body changes what an AI system can communicate. Gaze, posture, timing, proximity, and visual form can produce social presence, make identity salient, and alter how people coordinate with an agent. Across robots, virtual agents, and avatars, we study how those cues shape trust, identity, cooperation, and relationships over time.
How can embodied behavior be measured?
One strand of the lab examines interaction through movement. Motion capture, markerless pose estimation, egocentric video, and computer vision can show how bodies coordinate over time. Current work is developing measures of coupling, synchrony, and change across an interaction rather than treating a relationship as a single, static outcome.
Emerging directions
The lab is developing a broader account of interaction dynamics: how coordination and social expectations change through movement and repeated encounters. We are also asking what embodied AI design communicates about the person using it—whether a system treats that person as a partner, operator, or target. These are active research questions, not settled conclusions.
Methods and research environment
Our work combines controlled communication and HCI experiments with computer vision, XR, motion and pose data, simulation, and sensor-rich behavioral analysis. These tools support the research questions and remain secondary to them. “Kinesphere,” Rudolf Laban’s term for the space a body can reach, guides how we think about bodies acting within physical and virtual environments.
Selected evidence
Representative publications that ground the lab’s current research program.
Embodied movement and affect
MoEmo Vision Transformer combines movement and contextual information for emotion recognition in human–robot interaction.
Egocentric vision and body reconstruction
Fish2Mesh Transformer reconstructs 3D human meshes from egocentric fisheye video, supporting the study of embodied behavior from a first-person perspective.
Identity and social presence in XR
The Alt-Self examines inclusive avatar representation in social VR, while Social Presence in Human–Machine Communication provides a theoretical account of how machines are experienced as socially present.
Social kinematics and experimental methods
Work with virtual agents, proxemics, and representative experimental design provides the behavioral and methodological foundation for the lab’s emerging work on coordination over time.
Frequently asked
What is embodied human–AI interaction?
It is the study of how bodily and spatial cues shape communication between people and intelligent systems. Those cues may come from a physical robot, a virtual agent, or an avatar in an immersive environment.
What does the NTU Embodied Dynamics Lab study?
The lab asks how people interpret intelligent machines, how embodiment and repeated interaction shape social relationships, and how movement and multimodal behavior can be measured. It is based at NTU Singapore’s Wee Kim Wee School of Communication and Information.
Join or collaborate
I’m looking for PhD students and collaborators interested in embodied human–AI interaction, social presence, identity, cooperation, XR, computer vision, or behavioral measurement. Good fits may come from communication, HCI, computer vision, cognitive science, or design. Get in touch with a short note about the question you want to study.
For conversations about AI in higher education, curriculum, or external partnerships, see Technology, Education & Partnerships.
← Back to home