Date of Award

Fall 10-25-2010

Degree Type


Degree Name

Doctor of Philosophy (PhD)


Educational Psychology and Special Education

First Advisor

Dr. L. Juane Heflin, Ph.D.


Individuals with autism spectrum disorders (ASD) have documented sensory processing difficulties across the lifespan; however there is limited empirical support for the sensory-based interventions that have become ubiquitous with the population. This study was conducted to address this need and examine the effect of sensory-based interventions on skill acquisition for five elementary-age students with ASD. Proponents suggest that sensory-based interventions can be used to facilitate optimal levels of arousal so that children are available for learning. A single-case alternating treatments design was used to evaluate functional relations between the two sensory-based antecedent interventions and correct responding on expressive identification tasks. Upon visual analysis of the graphed data, functional relations were apparent for two participants. A positive relation between one sensory activity and correct responses was evident for a third student, but his rate of skill acquisition was too slow to verify a functional relation during the study. Results were undifferentiated for two students; one reached mastery criteria with both sensory-based interventions, while one made only modest improvement in expressive identification. Hierarchical linear modeling (HLM) was used to identify predictors of growth. Scrutiny of the results of the level-1 analysis revealed that there were significant differences among the participants at the start of the study (00 = 388.46, ²(4) = 45.97, p < .001) and that all of the students made significant gains during the study (10 = 2.35, t(4) = 3.43, p < .05). Using treatment as a predictor in Model 2 resulted in the finding of no significance for the sensory-based interventions in predicting growth. The two biggest level-2 predictors of student growth were age (11 = 0.055, t(2) = 6.403, p < .001) and IQ (22 = 0.21, t(2) = 13.41, p < .001). Although not clinically significant, Childhood Autism Rating Scale scores as a level-2 predictor of growth may have practical significance. Implications for mixed-modality research and applied practice are discussed.