| This investigation was undertaken to study the pause structure of oral discourse production following traumatic brain injury (TBI). Differing patterns of pause behavior would lend insight into deficient cognitive processes underlying discourse production. For this study, subjects included 40 TBI individuals and 40 normal controls who performed three discourse tasks: a picture description, a story retelling, and a moral formulation. The audio-taped discourse samples were orthographically transcribed. Filled pauses (FP) and silent (SP) pauses were then identified and measured via acoustical analysis.;The primary question regarding pause structure during oral discourse production was: Does the group membership and/or discourse task affect (1) pause frequency, (2) pause duration, and (3) pause location? Pause frequency and duration were found to differ between the TBI and control subject groups; however, no differences were found on measures of pause location. Further, the two groups showed a similar pattern of SP for all discourse tasks while the pattern of FP for the tasks differed.;Pause frequency, duration, and location were also found to differ across tasks but the pattern differed dependent on pause type (i.e., FP vs. SP). The frequency of FP increased across the three tasks while FP duration decreased. Frequency of SP was greatest for the story retelling task while SP duration was least for this task. In regards to pause locale, the proportion of pausing at syntactic boundaries decreased as the tasks increased in difficulty.;These differences between groups in pause frequency, duration, and use of filled pauses led to speculation regarding underlying cognitive processes for discourse production. Based on a proposed cognitive processing model, these differences indicate that breakdowns are occurring at the conceptual-processing stage for TBI subjects' oral discourse production. Since no differences were found between subject groups on pause location, the TBI subjects did not evidence difficulty at the semantic/syntactic level of processing. Further, differences in pause structure may result in listener perceptions of lack of speech naturalness and/or pragmatic problems in the discourse of TBI subjects. |
