Sensory integration of speech by a profoundly deaf subject using tactile aids

Previous research on tactual speech perception has focused on the relative contributions of lipreading and taction with normally hearing subjects. The integration of information from touch with that from lipreading and aided hearing by profoundly deaf subjects has not been investigated. In the present study, a profoundly deaf adult with previous tactual vocoder training [Lynch et al., J. Acoust. Soc. Am. Suppl. 1 82, S22 (1987)] was unable to identify a group of words in any one of three conditions, lipreading (L), aided hearing (H), or touch. The tactile devices used were a multichannel electrocutaneous vocoder (TV) and a two-channel vibrotactile aid (TA). To test sensory integration, the words were randomly assigned to one of seven conditions: (a) L + H, (b) H + TV, (c) H + TA, (d) L + TV, (e) L + TA, (f) L + TV + H, and (g) L + TA + H. Results indicate that: the subject integrated information across all three modalities, yielding highest performance in the L + TV + H and L + TA + H conditions; integration occurred when both lipreading and touch were used; integration occurred when both lipreading and aided hearing were used; and each of the tactile devices provided substantial speech information. Analyses of the subject's error patterns and sentence identification performance will also be presented. [Work supported by NIDRR, Dade County Public Schools, and NIH.]