Multimodal nonvocal nonword matching in adults who stutter

Significant data suggest that phonological working memory (PWM) is impaired in adults who stutter (e.g., Byrd, McGill, Usler, 2015; Byrd, Sheng, Bernstein Ratner, Gkalitsiou, 2015; Byrd, Vallely, Anderson, & Sussman, 2012; McGill, Sussman, Byrd, 2016). Some researchers have argued that the differences observed are driven by deficits in auditory processing as opposed to a core deficit in PWM (Hampton & Weber-Fox, 2008). This exclusive focus on auditory processing does not take into account results from studies that demonstrate PWM deficits in persons who stutter when auditory stimuli are not employed (e.g., McGill et al., 2016; Weber-Fox, Spencer, Spruill & Smith, 2004). No study has investigated PWM in persons who do and do not stutter on tasks that directly compare visual to auditory input performance differences. The purpose of this dissertation was to investigate PWM in adults who do and do not stutter using multimodal nonvocal nonword conditions including visual to visual, visual to auditory, and auditory to visual tasks. Data were coded for accuracy, manual reaction time, and error type. Analyses were conducted to evaluate the relationship between talker group, manual reaction time, accuracy, error types, and speed-accuracy tradeoff within and across conditions. Adults who stutter, compared to adults who do not stutter, demonstrated significantly decreased accuracy and increased manual reaction time on all nonvocal nonword conditions. Additionally, adults who stutter were significantly more likely to select a substitution foil compared to an accurate response than adults who do not stutter. Finally, neither adults who stutter nor adults who do not stutter demonstrated a speed-accuracy tradeoff. Results from the present study do not support the auditory exclusive argument posited in previous research. Regardless of input modality, PWM appears to be uniquely impaired for adults who stutter compared to adults who do not stutter.