Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.
Date of Graduation
5-12-2022
Semester of Graduation
Spring
Degree Name
Master of Science (MS)
Department
Department of Communication Sciences and Disorders
Second Advisor
Geralyn Timler
Third Advisor
Trevor Stokes
Fourth Advisor
Jaime Lee
Abstract
Research finds individuals with autism spectrum disorder (ASD) are relatively ineffective and/or inefficient at referential communication. However, this research typically uses static metrics of efficacy (how accurately messages were relayed) and efficiency (overall word count), rather than dynamic ones (e.g., Does the speaker alter subsequent descriptions when the listener previously misunderstood them?). The aim of this research is to use dynamic measures of efficacy and efficiency to examine how speakers with and without ASD adjust their message to meet listener needs across time. Fifteen older children with (n = 8) and without (n = 7) ASD were included. Participants interacted with two research assistants (RA1 and RA2). RA1 sat beside the participant. RA2 sat across from them, behind an opaque barrier. A board was positioned before the participant with a doll in its center and four black, ten-sided shapes surrounding the doll. RA1 explained that shapes were configured around RA2 the same way they were the doll. It was the participant’s task to use language to guide RA2 to select targeted shapes. There were 64 trials, and each trial was coded by the type of strategy. We created an “efficacy quotient” (EQ), which assigned 1s for trials when participants: 1) changed strategies when the previous trial was unsuccessful, 2) maintained strategies when it was successful. Remaining trials earned a 0. We summed 1s and divided this value by 63 (participants had opportunities to switch/maintain strategies from the second trial on). T-tests were used to compare EQs between groups. We measured the number of words within trials, where shorter utterances in later trials suggest referential shortening. We used linear modeling to compare slopes between groups. We found no significant difference in EQ between groups but found a significant effect of trial, indicating that both groups were equally effective at adjusting communication strategies based on previous success. However, participants with ASD were marginally more efficient: they were quicker to abbreviate utterances to reflect increased listener understanding. Findings emphasize the importance of examining interactions dynamically; such measures capture the realities of turn-taking and may identify strengths in autism that have heretofore gone unnoticed.
