Visual Attention to the Clinician's Face and Morpheme Acquisition during Conversational Recast Treatment
Author
Glickman, AlanaIssue Date
2019Keywords
AttentionConversational Recast Treatment
Developmental Language Disorder
Grammatical Morpheme Acquisition
Specific Language Impairment
Treatment
Advisor
Plante, Elena
Metadata
Show full item recordPublisher
The University of Arizona.Rights
Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction, presentation (such as public display or performance) of protected items is prohibited except with permission of the author.Abstract
Purpose. Enhanced Conversational Recast treatment is an input-based language therapy technique designed to help children with language disorders acquire missing grammatical morphemes in their speech. This version of conversational recast treatment requires clinicians to obtain the child’s attention before delivering each recast. This study examined the relationship between children’s looking behaviors in response to the clinician’s attentional cues and target morpheme acquisition. Method. Children received approximately 5 conversational recast treatment sessions per week for a total of 5 weeks. Progress was monitored through generalization probes assessing target morpheme use in untreated contexts. Video recordings of sessions were coded for children’s visual attention to their clinician’s face during the delivery of each treatment dose (i.e., recast). Reliability of coding was high. Results. Correlations between looking behaviors in response to attentional cues and performance on measures of generalization of morpheme use indicated a significant but negative association. Discussion. The results suggest that providing attentional cues prior to delivering treatment doses during conversational recast treatment may detract from the child’s attention to the relevant linguistic input, and decrease learning. Caution is warranted about this conclusion due to the possibility of clinician bias, and because coding from videos resulted in varying numbers of usable data points per session and per child.Type
textElectronic Thesis
Degree Name
M.S.Degree Level
mastersDegree Program
Graduate CollegeSpeech, Language, & Hearing Sciences