Effect of Avatar Head Movements on Communication Behavior and Subjective Evaluations of Presence and Success in Triadic Conversations.

All 16 participants were young adults aged 19 to 32 with normal hearing, so this says nothing about how people with hearing loss, older adults or children behave in the same situation. Only the participant was inside the virtual scene. The two conversation partners saw them only as video on a screen, which may have weakened any effect of the visual conditions on turn-taking. The conversation partners knew what the study was testing and the experimenter knew which condition was running, which may have changed how they moved, how loudly they spoke, and how turns were taken. The lengths of utterances and the timing of turns depend heavily on the settings used to detect when someone is speaking; other groups using different settings may get different numbers, and the authors say absolute turn-timing values cannot be compared across studies. The avatars had no facial expression, gestures or body movement, and the biggest differences all appeared in the live video condition, so head movement on its own may not be the feature that matters. The questionnaire could not detect the effects of head movement animation, so the lack of effect on how the conversation felt may mean the questions were not sensitive enough rather than that nothing changed. There were delays in the setup, about 512 ms for the streamed video and about 130 ms for the head movement animation, which the authors could not rule out as having affected the results.

Funded by the Deutsche Forschungsgemeinschaft (German Research Foundation), SFB 1330 project B1. The authors declared no conflicts of interest.

Do not read this as evidence that realistic avatar head movement helps people follow conversation in noise. Turn-taking gaps and overlaps were unchanged by the head movement animation, participants did not rate the conversation as any easier with more animation, and the biggest differences all came from the live video condition rather than from head movement. The study had 16 young normal-hearing volunteers and no patients, so it cannot tell you whether any of this helps a person with hearing loss.