How Counting, Representing, and Searching Do, and Do Not, Lessen Change Blindness for Person Substitutions

Participants in incidental change detection studies often miss large changes to central objects, such as substitutions of actors across cuts in videos. We investigated whether orienting tasks that induce specific forms of identity tracking, representation, and searching increase change detection. According to a minimal threshold hypothesis, incidental paradigms fail to elicit the level of processing necessary to trigger identity tracking. So, any orienting task that increases processing should increase change detection. In contrast, a selective processing account posits that multiple representation and comparison processes must be elicited independently to increase change detection. Although an orienting task might increase representation, it would not necessarily induce comparison as well and therefore fail to increase change detection. In four experiments we tested detection of actor substitutions when participants engaged in orienting tasks that required them to process actor identity but did not always require a specific combination of processes necessary to detect changes. Change blindness for actor substitutions persisted when participants counted the number of actors in the video and sometimes persisted when they were instructed to remember the changing actor for a subsequent lineup recognition test. Change blindness consistently diminished, however, when participants were shown the pre-change actor before or during the video and instructed to search for that actor in the video. Our results challenge the minimal threshold hypothesis by demonstrating that change blindness occurs even in tasks that require extensive processing of centrally attended changing objects. Our results suggest a selective processing account where identity tracking during real-world events involves a collection of processes that are applied selectively in a highly-task sensitive manner.