Context Matters! Social, Spatial, and Semantic Coordination with an Embodied Agent in Immersive Analytics

In plain English

A design framework rather than an experiment. The authors argue that an embodied agent sharing physical space with a user has to coordinate on three kinds of context — spatial, social and semantic — each of which splits into how the agent is situated and what it senses and reflects back to the user. They illustrate it with Cobot, a HoloLens 2 agent using gaze cues, spatial audio, proxemics-based navigation at roughly 80 cm, and a head-tilt "thinking" cue. There are no participants and no measurements, only informal testing, so treat it as vocabulary for design reviews rather than evidence; the value is a checklist of the contextual behaviours that make a co-present agent read as competent.