Consequences of Fractal Event Perception: Lessons from Real and Reel Events
Date of Completion
As we perceive and act in our environment, our experience is organized into events. Although events have often been treated as sequential divisions of time, an alternative is pursued here: Perception of events is nested rather than sequential, and this nested temporal structure is fractal. The nature of the temporal structure of both natural and cinematic events is investigated in five experiments using an event segmentation methodology in which observers tap to event boundaries or those boundaries are extracted from the cuts imposed by film editors. Temporal structure in all experiments is fractal, as indexed by the Hurst exponent H. Htap, does not differ for natural settings that differ in activity level (Experiment 1) but it is affected by observer characteristics such as expertise and anxiety (Experiment 2). H serves as an index of learning or attunement (Experiment 3), both for film editors (via Hedit) and for observers (via Htap) encountering a novel film-event structure. When the coordination between Hedit and the source event is broken by re-editing a movie so as to produce two versions with differing values of Hedit, Htap reflects competition between two attractors, the original H and the new Hedit (Experiment 4). Hedit has consequences for viewers' understanding of and memory for content in the two edited versions of that film (Experiment 5). Taken together, these experiments encourage consideration of film as a medium, and segmentation—coupled to fractal analysis—as a methodology for studying cognitive constraints on event perception. ^
Blau, Julia James Carroll, "Consequences of Fractal Event Perception: Lessons from Real and Reel Events" (2011). Doctoral Dissertations. Paper AAI3485235.