A Comparison of Speech Onset Latencies Between Persons Who Stutter and Persons Who Do Not Stutter Across Varied Phonological Priming Conditions

Abstract

A recent theory of stuttering, the "Covert Repair Hypothesis of Disfluencies" (Kolk & Postma, in press; Postma & Kolk, 1992, 1993), accounts for the difference between persons who stutter (PWS) and persons who do not stutter (PWNS) by concluding that PWS are slower than PWNS in their phonological encoding abilities. This belief is supported through experimental studies by Bosshardt (1990) and Postma et al (1990), both of which found PWS to be slower than PWNS in silent reading tasks. In addition, Wijnen and Boers (1994) found that PWS demonstrate longer speech onset latencies than PWNS at baseline, but then approximate the times of PWNS upon "phonological priming." They interpreted their results to indicate that in PWS "the encoding of noninitial parts of syllables, particularly the (stressed) vowel, is delayed" (p.1). The purpose of the present study was to test the covert repair hypothesis, as it is applied to the difference between PWS and PWNS, while eliminating some of the potential biases found in earlier studies. The research question for this study was: "Is there a significant difference in speech onset latencies between PWS and PWNS across three conditions in which CV, VC, or no part of a one syllable (CVC) word is primed prior to naming of the target?" Six PWS ages 27 to 47 were recruited from both the Portland State University Speech and Hearing Clinic and a stuttering support group that meets on campus. All PWS were native speakers of English and diagnosed as a PWS by an ASHA certified speech-language pathologist (SLP). In addition, all PWS still considered themselves to be a PWS through self-report/interview. Only one of the six PWS was currently receiving treatment. The control group consisted of 20 PWNS ages 18 to 37 recruited from Portland State University. All PWNS were native speakers of English. All subjects performed a picture naming task designed to measure speech onset latencies across varied phonological priming contexts. Subjects were tested individually by being seated in front of a computer monitor and naming line drawings of common objects as they appeared on the screen. Subjects were asked to name the pictures as quickly and as accurately as possible. The task consisted of 504 experimental trials, presented in two blocks of 252 trials. Following completion of the task, all naming errors, apparatus malfunctions, and extreme outliers were omitted prior to statistical analysis. Mean speech onset latencies of the two groups