Abstract
Research has shown that the behavior of a bystander can influence decisions about whether or not to evacuate a room (Kinateder & Warren, 2016). Here, we investigated how neighboring pedestrian walking speeds influence exit choice and exit time using immersive virtual reality (VR), as well as a matched real-world condition for validation purposes. Forty-five participants performed a simple egress task in which they exited a room via one of two doors. Two other pedestrians were also present (avatars in VR and confederates in the real world). On each trial, the “variable” pedestrian walked towards one door by traveling at one of three average speeds (1.0m/s, 1.5m/s, or 2.0m/s), and the “constant” pedestrian walked towards the other door at a speed of 1.5m/s. Pedestrian exit door was counterbalanced, yielding six randomly presented trials in VR and six matched trials in the real world. The VR/real-world condition order was counterbalanced across participants. To start each trial, the two pedestrians began walking to their assigned doorway at their assigned speeds. Then, the participant was cued to exit the room. After the participant exited, all parties were instructed to return to the center of the room to begin the next trial. Results indicated a general bias to follow the faster pedestrian. Further, the variable pedestrian speed also significantly influenced exit time such that participants exited faster during the 2.0m/s trials and slower during the 1.0m/s trials. Effects were similar for VR and real-world conditions, validating the use of VR in this context. Together, these results suggest that bottom-up visual motion cues biased exit choice and exit time. The influence of top-down cognitive processes via individual strategy differences will also be explored. Future work will employ more goal-directed evacuation scenarios.