Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


College of Humanities and Social Sciences


Communication Sciences and Disorders

Thesis Sponsor/Dissertation Chair/Project Chair

Mary Boyle

Committee Member

Sarita Eisenberg

Committee Member

Michael Boyle

Committee Member

Elizabeth Armstrong


Casual conversation is essential to everyday life. Because of their language impairment, most people with stroke-induced aphasia have trouble engaging in social interactions and may, as a result, experience a decrease in psychosocial well-being. There is a great need, therefore, to identify the interventions most likely to improve the everyday talk of individuals affected by aphasia. Approaches to intervention that aim to improve the social participation and communicative success of people with aphasia during real-life interactions, known as conversation-based treatment, have increased in popularity.

When delivered in a group setting, conversation-based treatments have been shown to increase scores on standardized aphasia assessments and improve narrative discourse production. If and how conversation-based group treatment may change the casual conversation of persons with aphasia is not well understood. The identification of suitable discourse outcome measures has represented a significant hurdle in the search for post-intervention changes to conversational discourse production.

The present study sought to determine the reliability of the speech function network and its potential utility as a discourse outcome measure following conversation-based group treatment. Pre- and post-treatment conversational discourse samples collected from 22 participants with chronic aphasia were coded for moves within the speech function network. We determined that the test-retest reliability of speech function analysis was sufficient for use in a group treatment study; however, inter-rater reliability ranged from poor to excellent for different moves within the network.

Post-intervention changes in moves within the speech function network were also examined. The results of this analysis indicate that participation in conversation-based group treatment did not result in statistically significant changes in the moves used by the study participants during conversational discourse. Patterns of behaviors identified through the speech function analysis broaden our understanding of how dyads affected by aphasia engage in everyday talk.

File Format