Perceptual Dimensions of Visual Scenes Relevant for Simulating Low-Altitude Flight
Abstract
Multidimensional scaling was used to identify the features of real- world terrain that are salient to pilots during low-altitude flight. The subjects were pilots experienced flying in the Southwest United States (Experiment 1) and pilots experienced flying in Europe (Experiment 2). The stimuli were videotape segments (Dynamic Presentation) and still photographs (Static Presentation) depicting low-altitude flight over a variety of real-world terrains. Pilots rated pairs of terrains with respect to similarity of visual cues for low-altitude flight. Terrains were also rated on eight bipolar rating scales representing a variety of terrain characteristics thought to be of possible relevance to pilots. similarity ratings were submitted to a multidimensional scaling analysis using the procedure ALSCAL. Two-dimensional solutions were deemed most appropriate in all cases. Bipolar ratings were submitted to a multiple regression analysis in which ratings on each scale were regressed over dimensional coordinates. Results of Experiment 1, Dynamic Presentation, revealed dimensions corresponding to (1) terrain contour, and (2) object size and spacing. Results for Static Presentation were less interpretable suggesting the possibility of a single dimension capturing the presence/absence of global scene detail. In Experiment 2, results for both presentation modes replicated Experiment 1, Dynamic Presentation, although the fit of the data remained superior with Dynamic Presentation. Taken together, these results provide consistent evidence that pilots flying at low altitudes perceive variation in terrain contour and object size and spacing.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 1992
- Accession Number
- ADA254645
Entities
People
- James A. Kleiss
Organizations
- University of Dayton