Immersive collaborative virtual environments (CVEs) allow us to interact with geographically distant others while experiencing social presence to a degree that goes far beyond text chatting or teleconferencing. Moreover, these environments provide this high level of realism within social contexts that are impossible in the physical world. Given these facts, CVEs provide behavioral researchers with an ideal platform to study social interaction both within and between geographically and culturally distinct communities. The study reported here leveraged these two unique capabilities of CVEs within a persuasive context by: (1) placing people who are seated in physically distal places into the exact same virtual world, and (2) structuring virtual space to maximize persuasion. Specifically, we report data from a study in which pairs of participants listened to a speaker deliver a persuasive passage within the same digital immersive virtual room. The individual members of each pair were separated by hundreds of kilometers, located at two different college campuses. Within the CVE, we digitally transformed the placement of participants' seats in the virtual classroom. Participants in the front of the classroom were more persuaded by the speaker and had more positive impressions of the speaker. Patterns in both persuasion and memory differed between campuses. Together these findings speak to the utility of wide range CVEs to maximize persuasion and demonstrate the viability of using CVEs for inter-site research.

The current study investigated the value of using immersive virtual environment technology as a tool for assessing eyewitness identification. Participants witnessed a staged crime and then examined sequential lineups within immersive virtual environments that contained 3D virtual busts of the suspect and six distractors. Participants either had unlimited viewpoints of the busts in terms of angle and distance, or a unitary view at only a single angle and distance. Furthermore, participants either were allowed to choose the angle and distance of the viewpoints they received, or were given viewpoints without choice. Results demonstrated that unlimited viewpoints improved accuracy in suspect-present lineups but not in suspect-absent lineups. Furthermore, across conditions, post-hoc measurements demonstrated that when the chosen view of the suspect during the lineup was similar to the view during the staged crime in terms of distance, accuracy improved. Finally, participants were more accurate in suspect-absent lineups than in suspect-present lineups. Implications of the findings in terms of theories of eyewitness testimony are discussed, as well as the value of using virtual lineups that elicit high levels of presence in the field. We conclude that digital avatars of higher fidelity may be necessary before actually implementing virtual lineups.