Abstract
Change blindness (CB) occurs when observers fail to notice changes that occur from one view to the next. CB has been demonstrated using many different stimulus types (shapes, pictures, movies, real objects) in a many different contexts (from laboratory to “on the street”). The current experiment utilized virtual reality (VR) to explore incidental detection of changes that could not easily happen in the real world. Participants (N = 73) played a version of a “whack-a-mole” game in a VR arcade. Before or after playing the game, text on the front wall of the arcade instructed participants to look at the posters on the walls on the left and right side of the room. The instructions ensured each wall was looked at twice, and head rotation was tracked to ensure participants complied. Unbeknownst to participants, different attributes on the left and/or right wall changed across views: one set of posters changed in spatial configuration and one wall changed color. Changes occurred while participants looked in the direction opposite of the changing feature, ensuring the changes would not capture attention. Participants were later asked a series of yes/no questions about which objects in the virtual environment had changed. Correcting for false alarms, only about 15% of changes were detected. Whether the changes occurred on the same or different sides of the room did not matter. Hit and false alarm rates were similar for posters changing configuration and walls changing color. Moreover, detection of one change was independent of detecting the other. Pre- and post-change looking duration predicted detection of wall color changes (weakly), but not poster configuration changes. Overall, the results are inconsistent with the idea that spatial configuration changes are better detected than other visual changes, but consistent with the idea that different features of the visual world are represented independently.