The effects of similarity neighborhood structure on auditory word recognition
Paul A. Luce · The Journal of the Acoustical Society of America · 1986
The present study examined how the structure of similarity neighborhoods of spoken words affects auditory word recognition. Similarity neighborhoods were computed on phonetic transcriptions of approximately 20 000 words in an on-line version of Webster's pocket dictionary. Among the variables of interest were: (1) the density of the similarity neighborhood; (2) the frequencies of words within the neighborhood; and (3) the frequency of the target word itself. Nine-hundred and twenty-eight three-phoneme, monosyllabic words varying on each of these variables were presented at each of three signal-to-noise ratios for identification. In addition, confusion matrices were obtained for all initial and final consonants and vowels occurring in the 928 stimuli in order to account for basic differences in segmental confusability and intelligibility. The results will be discussed in terms of the combined effects of similarity neighborhood structure and segmental confusability on auditory word recognition. [Work supported by NIH grant No. NS-12179 to Indiana University.]