Reliability and Validity of a Sensory Seeking Scale in the Sensory Processing Assessment

Background: Sensory processing abnormalities have been found highly prevalent, although not universal, among young children with Autism Spectrum Disorder (ASD). The Sensory Processing Assessment (SPA; Baranek, 1999b) is a 20 minute semi-structured, play based observational tool designed to measure hyperresponsiveness (an exaggerated response to sensory stimuli) and hyporesponsiveness (a lack of response to sensory stimuli). Sensory seeking, a third construct of sensory processing, has been demonstrated by recent factor analytic models. Thus, a sensory seeking scale was added to the SPA. Previously untapped by observational measurement, sensory seeking is characterized by initiation and preoccupation with oneself, an activity, or object, and can provide strong visual, auditory, vestibular, tactile, proprioceptive, olfactory, or gustatory input. Engagement may be prolonged, repetitive, or stereotypical in its presentation, although repetition is not a sufficient characterization for sensory seeking.

Objectives: The purposes of this study were to measure the (a) inter-rater reliability and (b) “known-groups” discriminative validity of a sensory seeking scale in the Sensory Processing Assessment.

Methods: Construction of the sensory seeking scale drew from existing literature and clinical observations of behaviors related to sensory processing.  The sensory seeking scale includes 8 items rated on a scale from 0 (none observed) to 2 (intense and frequent). The current sample included 52 children, aged 7 to 95 months, with ASD, developmentally delay (DD), or typical development (TYP). (Note: Total sample size will be 105 by May 2010). An ANCOVA was conducted to evaluate the relationship between diagnostic group (ASD, DD, TYP) and sensory seeking scores, while controlling for mental age.

Results:  Preliminary data analysis revealed inter-rater reliability of the sensory seeking scale is ICC=.95. Results of the ANCOVA found significant main effects for group (F= 8.482, p<001). Follow up Tukey’s LSD showed significant differences between the ASD group, and the DD and TYP groups, p<.02 and p<.001 respectively.

Conclusions: These preliminary findings suggest that the sensory seeking scale of the Sensory Processing Assessment is a reliable tool for measuring behaviors previously untapped by an observational assessment, with excellent inter-rater reliability. The sensory seeking scale distinguished the ASD group from the developmentally delayed and typically developing groups, suggesting that high levels of sensory seeking are unique to ASD, and a possible discriminating feature of the disorder. There are mixed findings suggesting that repetitive behaviors, often conflated with sensory seeking, distinguish DD from ASD. Follow up analyses will be completed to determine if specific types of sensory seeking behaviors are more likely to discriminate groups in order to further elucidate the sensory phenotype of children with ASD.