The perceptual structure of multidimensional input device selection

Concepts such as the logical device, taxonomies, and other descriptive frameworks have improved understanding of input devices but ignored or else treated informally their pragmatic qualities, which are fundamental to selection of input devices for tasks. We seek the greater leverage of a predictive theoretical framework by basing our investigation of three-dimensional vs. two-dimensional input devices on Garner's theory of processing of perceptual structure in multidimensional tasks. Two three-dimensional tasks may seem equivalent, but if they involve different types of perceptual spaces, they should be assigned correspondingly different input devices. Our experiment supports this hypothesis and thus both indicates when to use three-dimensional input devices and gives credence to our theoretical basis for this indication.