Vibratory temporal integration as a function of pattern discriminability

Abstract

In a recent study, we measured temporal integration for vibrotactile spatial patterns presented to the fingertips. The patterns were generated on the tactile array of the Optaeon, a reading aid for the blind (Bliss, Katcher, Rogers, & Shepard, 1970;Craig, 1980). Several different tasks involving letter recognition were used in the study. In one task, letters were divided in half and the time between the presentation of the first and second half was varied, with the result that recognition performance declined as the time between the onset of letter halves increased. In a second task, the time between the onset of a letter and the onset of its complement, that portion of the tactile array not occupied by the letter, was varied. Letter recognition improved as the letter emerged from its complement (Craig, 1982). The results were similar to previous measures involving detection of vibratory stimuli in that the major portion of the integration took place within 50 to 100 msec of the onset of the first pattern (Gescheider, 1976;Green, 1976;Verrillo, 1965).The measurements with patterned stimuli and the temporal integration functions thus generated were similar to one another in that the skin showed complete temporal integration only over time periods of 10 msec and performance reached asymptotic levels as the time between stimuli approached 100 msec. There were, however, some differences in the functions produced by the several tasks, and it is the differences between functions that are the focus of this note.The task involving a letter and its complement produced sharper temporal integration functions, that is, performance reached asymptotic levels at briefer temporal separations than in the task in which letters were divided in half. The letter-complement task also turned out to be an easier task, in that levels of performance were higher than in the divided-letter task. The results suggested either that differences in the nature of the two tasks (such as ceiling effects) may have resulted in differences in the temporal integration functions or that the ease of discriminability of spatial patterns might produce differences in the functions, that is, patterns that

Vibrotactile pattern discrimination and communality at several body sites

“…These restrictive processes operate regardless of whether the pattern set is presented in a coherent display like a vibrotactile matrix, or one that limits the spatial integration of the patterns, as occurs on the separate fingers of the hand or over the body (see, e.g., Craig, 1984;Craig et al, 1985). It is also possible that such limitations may playa role in the discrimination of letter-like patterns (Craig, 1984;Evans, 1987;Horner, 1991Horner, , 1992. Such factors may be central to why most users of tactile aids report slow rates in processing of tactile information.…”

“…The three body loci will be tested with vibrotactile arrays specifically designed to match the spatial acuity of each (i.e., with varied contactor spacing), allowing examination of the sites under equivalent conditions (see, e.g., Cholewiak & Craig, 1984;Daley & Singer, 1975). In these arrays, the contactor separation is only slightly larger than the error of localization on the corresponding site.…”

Vibrotactile masking: The role of response competition

“…Several experimental paradigms can be generated from the situation in which there are two patterns presented sequentially, and there are either one or two sites ofstimulation. The two patterns may be sequentially presented to the same site, and the subject required to combine them-a temporal integration task (Craig, 1982a(Craig, , 1984. The two patterns may be presented to separate sites, and the subject required to combine or compare the two patterns-a divided attention task (Craig, 1985).…”

The impact and specificity of nerve perturbation on novel vibrotactile sensory letter learning

“…The delivery of tactile information for abstract No paresthesia acquisition or transfer concepts has been found to be difficult. Craig (2000) found an accuracy rate of 30-60% for vibrotactile letter learning, while likewise Foulke and Broadbeck (1968) found a 40.4% error rate in teaching Morse code letters using electrocutaneous stimulation. Tactile task difficulty is one reason why tactile interfaces that require complex information processing are sparse.…”

“…The accuracy results from the present study are in agreement with previously established accuracy data regarding tactile letter acquisition. For example, Craig (2000) found accuracy performance in the range of 30-60% correct when vibrotactile letters were presented. Foulke and Broadbeck (1968) conducted a study that also found poor performance accuracy when they attempted to use electrocutaneous stimuli to deliver Morse code to participants at an isolated location.…”