Length, grammatical complexity, and rate differences in stuttered and fluent conversational utterances of children who stutter

The purpose of this study was to assess length, grammatical complexity, and articulatory speaking rate differences in stuttered and perceptibly fluent conversational utterances produced by children who stutter. Subjects were 15 boys who stutter (mean age = 51.2 months, SD = 8.09 months), each of whom was audiotaped and videotaped while interacting with his mother during a 30-min play/conversation period. Twenty-five stuttered and 25 perceptibly fluent utterances from each subject's conversational speech sample were measured in terms of syllabic length, grammatical complexity, and articulatory speaking rate, and each utterance was categorized as “high” or “low” in length, grammatical complexity, and articulatory speaking rate relative to each subject's median for each of the three variables. Results indicated that syllabic length for stuttered utterances was significantly greater than that for perceptibly fluent utterances. Additional analysis showed that significantly more stuttered utterances were categorized as “high” in length and/or grammatical complexity, and that significantly more perceptibly fluent utterances were categorized as “low” in length and/or grammatical complexity. Results supported neither the notion that articulatory speaking rate differs between stuttered and perceptibly fluent utterances, nor the idea that articulatory speaking rate, when considered together with either utterance length or utterance grammatical complexity, determines whether an utterance will be stuttered or perceptibly fluent. Results are discussed in terms of current speech production models as well as current theory that suggests that speech fluency breakdowns are more likely to occur in utterances for which task demands (performance) exceed an individual's typical level of performance (capacity or ability).