Title

Bimanual haptic attention

Date of Completion

January 1999

Keywords

Psychology, Experimental

Degree

Ph.D.

Abstract

Humans are capable of performing a variety of tasks simultaneously. Some of the most common—walking, carrying objects, using tools—are controlled by the haptic perceptual system. The ability to perform multiple tasks simultaneously is classically referred to as selective attention. Typically applied to vision and hearing, attention theories do not seem to be applicable to the range of simultaneous haptic activities people are able to perform. Five experiments were designed to attempt to bridge the gap between the classical theories of attention and recent work on attention in the haptic domain. In particular, two methodological changes to the standard touch paradigm—increasing the number of distractors and overlaying a simultaneous, second task—provide more classical attentional settings. The primary task was perceptual: judge either the whole or partial length of the attended rod; the secondary task was rhythmic: wield the rods in both hands. Experiment 1 found that perception of whole length was generally unaffected by these changes. Experiment 2, however, showed that perception of partial length, which already requires a degree of selectivity, was undermined both in its accuracy and in its dependence on rotational inertia—the usual physical constraint on perceived length. Experiments 3 and 4 reinforced this disparity between perceiving the two properties and allowed the possibility that perceived partial length may have become dependent on perceived whole length. Experiment 5 evaluated this possibility formally and found a deterioration in perceived partial length that was nonetheless perceptually independent of perceived whole length. Discussion focused on the classical theories of attention and how they may be adjusted to explain the tasks controlled by the haptic system. Attention as an active selection of invariants of a structured energy array was described as an alternative. ^

COinS