The effects of contextualization on fluency in three groups of children

This study investigated the effects of contextualization on fluency in 12 school-age children who stutter (CWS), 11 children with language impairment (CLI), and 12 children with normally developing fluency skills (CNF). Participants in the study were between the ages of 8 and 12 years and were matched for age and sex. Four discourse samples were elicited by asking participants to (a) generate two scripts related to cooking and (b) retell two stories. Having objects or pictures immediately available contextualized a cooking task and a retelling task; another set of cooking and retelling tasks were decontextualized. Moments of disfluency were identified and coded for three primary categories of disfluency: stuttering-type, normal-type, and mazing. For CWS, a significant reduction in frequency of stuttering was noted in the contextualized script generation, and mazing occurred at a significantly higher frequency than did stuttering-type or normal-type disfluencies across the four tasks. For all three groups, both decontextualized conditions produced greater frequencies of normal-type disfluency and mazing. In addition, narrative retelling tasks yielded higher frequencies of disfluency than did the two cooking scripts.