Participants in the intervention group were provided with three options for interacting with the computer: using a static male physician avatar, a female physician avatar (photographic images of a provider conversing as if with the patient) or no avatar (text only).