Abstract
During reading, eye-movements are closely related to comprehension, but the opposite has been shown in film viewing. Recent studies have shown that the high attentional synchrony found during film viewing may leave little room for comprehension based eye-movement differences. We therefore tested the following two competing hypotheses: 1) The Cognitive Task Hypothesis: high-level task processes guide eye-movements, versus 2) The Tyranny of Film Hypothesis: movies produce strong attentional synchrony regardless of higher-level processes. Previous experiments testing similar hypotheses showed that manipulations of viewer comprehension showed little effect on eye-movements, thus strongly supporting the tyranny of film hypothesis. To test whether the tyranny of film could be "turned-off," thus supporting the Cognitive Task hypothesis, participants were explicitly told their task was to draw a detailed map from memory of all locations depicted in a video, the 3-minute opening shot from "Touch of Evil" (Welles, 1958). In this clip a bomb is placed in a car at the beginning, and is shown until the bomb in the car is just about to explode. Map task viewers' eye-movements were compared to participants who watched the same clip while free viewing. To measure viewers' comprehension at the end of the clip, we asked them what would happen next, which showed the large expected task difference in terms of mentioning the bomb. The free-view group was much more likely to mention the bomb, indicating they were following the narrative, and are thus referred to as the Comprehension condition. For the eye-movement measures, compared to the Comprehension condition, the Map Task produced significantly longer saccades, different fixation patterns, and less looks at the car with the bomb. Thus, while comprehension differences produce only small effects in film viewers' eye-movements, a task implicitly at odds with comprehending the film narrative counteracts the tyranny of film.
Meeting abstract presented at VSS 2016