How Do Viewers Spontaneously Segment Animated Diagrams of Mechanical and Biological Subject Matter?

Abstract

A challenges for learning from animated diagrams is to first parse the continuous flow of information into discrete event units. Inadequacies in this parsing process can prejudice the quality of the mental model constructed from the depiction. One approach that has been proposed for ameliorating such problems is for the designer to pre-segment the animation. However, the pre-segmentation techniques used tend to be either intuitive or based on an expert’s understanding of the subject matter. Neither of these approaches takes proper account of the psychological processing that must occur for an external animation to be properly internalized. This poster reports a study of the processes that learners spontaneously use when asked to segment whole animations into events. It compared segmentation of two contrasting diagram types, one representing a mechanical system and the other a biological system. The number of events identified was low relative to the number that were actually present. There were deficiencies in participants’ placement of event boundaries and in their characterization of inter-event relationships. Identification of events in the mechanical system proceeded from micro to macro, this order was reversed with the biological system.