The effects of size changes on haptic object recognition.

Two experiments examined the effects of size changes on haptic object recognition. In Experiment 1, participants named one of three exemplars (a standard-size-and-shape, different-size, or different-shape exemplar) of 36 categories of real, familiar objects. They then performed an old/new recognition task on the basis of object identity for the standard exemplars of all 36 objects. Half of the participants performed both blocks visually; the other half performed both blocks haptically. The participants were able to efficiently name unusually sized objects haptically, consistent with previous findings of good recognition of small-scale models of stimuli (Lawson, in press). However, performance was impaired for both visual and haptic old/new recognition when objects changed size or shape between blocks. In Experiment 2, participants performed a short-term haptic shape-matching task using 3-D plastic models of familiar objects, and as in Experiment 1, a cost emerged for ignoring the irrelevant size change. Like its visual counterpart, haptic object recognition incurs a significant but modest cost for generalizing across size changes.