Academic journal article Perception and Psychophysics

Multisensory Numerosity Judgments for Visual and Tactile Stimuli

Academic journal article Perception and Psychophysics

Multisensory Numerosity Judgments for Visual and Tactile Stimuli

Article excerpt

To date, numerosity judgments have been studied only under conditions of unimodal stimulus presentation. It is therefore unclear whether the same limitations on correctly reporting the number of unimodal visual or tactile stimuli presented in a display might be expected under conditions in which participants have to count stimuli presented simultaneously in two or more different sensory modalities. In Experiment 1, we investigated numerosity judgments using both unimodal and bimodal displays consisting of one to six vibrotactile stimuli (presented over the body surface) and one to six visual stimuli (seen on the body via mirror reflection). Participants had to count the number of stimuli regardless of their modality of presentation. Bimodal numerosity judgments were significantly less accurate than predicted on the basis of an independent modality-specific resources account, thus showing that numerosity judgments might rely on a unitary amodal system instead. The results of a second experiment demonstrated that divided attention costs could not account for the poor performance in the bimodal conditions of Experiment 1. We discuss these results in relation to current theories of cross-modal integration and to the cognitive resources and/or common higher order spatial representations possibly accessed by both visual and tactile stimuli.

The majority of previous visual numerosity judgment studies have reported a difference in the accuracy and latency of people's enumeration responses to presentations of small versus large numbers of items (see, e.g., Atkinson, Campbell, & Francis, 1976; Jevons, 1871; Trick & Pylyshyn, 1993,1994; Weiss, 1965). When the number of items presented is small (typically between one and four), the items appear to be processed very rapidly and nearly without error (see, e.g., Atkinson, Campbell, & Francis, 1976). Increasing the number of items presented to more than four typically produces a large increase in both average response latencies and error rates, often giving rise to a discontinuity in the slope of the response latency and error functions. Such results have been interpreted by many authors (e.g., Jensen, Reese, & Reese, 1950; Kaufman, Lord, Reese, & Volkmann, 1949; Mandler & Shebo, 1982; Peterson & Simon, 2000; Trick & Pylyshyn, 1993; Wender & Rothkegel, 2000) as providing evidence for the existence of two qualitatively different enumeration processes: subitizing, which is specialized for small numbers of items, and counting, which is specialized for larger numbers of items. Subitizing is fast, accurate, and preattentive, whereas counting tends to be slow, error prone, and attention demanding.

Numerosity judgments have been studied outside of the visual modality as well. For example, ten Hoopen and Vos (1979) demonstrated that a distinction between subitizing and counting can be observed when people enumerate auditory stimuli (see also Cheatham & White, 1954; Garner, 1951;Taubman, 1950; White & Cheatham, 1959). However, the fact that the stimuli in ten Hoopen and Vos's research were presented sequentially makes any direct comparison with the results of previous visual numerosity judgment studies, in which the stimuli were nearly always presented simultaneously (though see Hill, 1971; Lechelt, 1975; andViviani, 1979, for exceptions), difficult. Kashino and Hirahara (1996) presented different numbers of voices simultaneously and found that the accuracy of participants' estimates of the number of speakers was nearly perfect for up to two talkers but deteriorated substantially whenever three or more talkers were presented.

Finally, numerosity judgments have also been studied within the tactile modality. For example, Gallace, Tan, and Spence (2006c) found that the accuracy of participants' numerosity judgments when counting up to seven simultaneous vibrotactile stimuli distributed across their body surface decreased linearly as the number of stimuli increased, with no clear distinction between counting and subitizing evident (see also Lechelt, 1974, 1975, for the results of tactile serial numerosity judgment experiments). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.