Effects of Interactivity and Spatial Ability on the Comprehension of Spatial Relations in a 3D Computer Visualization Department of Psychology, UCSB, Building 551 Room 1332, Santa Barbara, CA 93106-9660 USA Department of Geography, UCSB, Ellison Hall Room 3611, Santa Barbara, CA 93106-4060 USA High Spatial Low Spatial Active Passive Figure 1: Performance on the cross-section drawing task by interactivity condition, as a function of spatial ability. Results Cross-Section Task (Proportion Correct) This experiment was designed to investigate the roles of interactivity and spatial visualization ability in the comprehension of 3D computer visualizations. Undergraduates were presented with a fictitious anatomy- like structure in the form of both printed 2D images and a 3D computer visualization that could be rotated in x, y and z dimensions. A superimposed vertical or horizontal line on the printed images indicated where they should imagine the structure had been sliced. The task was to draw the cross- section at that point. The drawings were assessed for spatial understanding using a standardized scoring scheme. Sixty participants were randomly allocated to one of two conditions. The active group was allowed to rotate the computer visualization at will via keyboard controls during the drawing task. The passive group had no control over the movements. Using a yoked pairs design, the manipulations performed by the active participants were recorded and later played back to the passive participants, so that both members of each pair received the same visual information. Spatial ability was measured via the Mental Rotation Test (Vandenberg & Kuse, 1978) and a modified version of Guay’s Visualization of Views test (Eliot & Smith, 1983). condition, i.e. when participants were allowed to manipulate the 3D model, the performance means of high and low spatial individuals were brought closer together. While low- spatial participants were helped by interactivity, this benefit did not extend to high-spatial individuals. We are currently undertaking a replication study with a more intuitive control mechanism, to establish whether these findings arose from the nature of the interface or from interactivity per se. Proportion correct Introduction and Method Cross-Section Task (Proportion Correct) Madeleine Keehner 1 (keehner@psych.ucsb.edu) Daniel R Montello 2 (montello@geog.ucsb.edu) Mary Hegarty 1 (hegarty@psych.ucsb.edu) Cheryl Cohen 1 (cacohen@umail.ucsb.edu) There was no main effect of condition, indicating no significant difference between the active and passive control conditions. However, a main effect of spatial ability was found (median split; F=9.38, p .1; Figure 1). In line with this finding, the correlation between spatial ability and performance was relatively attenuated under active control (r=.29, p>.1), compared to passive viewing (r=.51, p<.005; Figure 2). This research was supported by NSF grant # EIA - 0313237. Discussion References The data indicate that having active control of the computer visualization did not benefit overall performance. A more important predictor of success was individual differences in spatial ability. However, the contribution of this factor was stronger in the passive condition than in the active Vandenberg, S. G., & Kuse, A. R. (1978). Mental rotations, a group test of three-dimensional spatial visualization. Perceptual & Motor Skills, 47, 599-604. Eliot, J., & Smith, I M. (1983). An international directory of spatial tests. Windsor, Berkshire: NFER-Nelson. Rsq = 0.0854 0.0 Spatial Ability (z-scores) Active Group Rsq = 0.2566 Spatial Ability (z-scores) Passive Group Figure 2: Correlation between spatial ability and performance, by interactivity condition. Acknowledgments