Differential tolerance for the judgement of dysfluencies as stuttered speech in native and non-native speech-language pathologists.

PURPOSE: Accurate diagnosis of stuttering requires careful observation by trained professionals. Speech-language pathologists (SLPs) typically identify stuttering using a perceptual judgement. However, it is unclear if this perception is based on the amount or the type of dysfluencies. Hence, the study aimed to determine the frequency of syllable/part-word repetitions judged as stuttered and to determine if the listener's language influences their judgement of fluency.

METHOD: Narrative samples of a Kannada-speaking child who stutters containing 1 %, 3 %, 5 %, 10 %, and 15 % of syllable or part-word repetitions were judged by 27 Kannada and Malayalam-speaking SLPs each as fluent, disfluent, or stuttered. Further, an opinion about recommending the child for speech therapy was taken. The chi-square test and Fisher's exact test were performed to compare the difference between the two groups.

RESULTS: Comparison between the five frequencies of dysfluencies and the judgement of speech revealed a significant difference for Kannada and Malayalam language groups. The non-native SLPs required more dysfluencies to judge speech as stuttered compared to the native SLPs. Both groups of SLPs recommended speech therapy for samples above 5 % dysfluencies.

CONCLUSIONS: Familiarity of Malayalam-speaking SLPs with the Kannada language and the experience of the participants in the assessment of stuttering could have influenced the differences obtained in this study. However, current results indicate a differential threshold for the perception of dysfluencies (syllable and part-word repetitions) and the influence of the listener's language familiarity. Hence, language familiarity may be a factor in the judgement of dysfluencies as fluent, disfluent, and stuttered speech.