What’s That: Whole Object and Taxonomic Constraint in Children with Autism
Files
Related Publication Link
External Link to Data Files
Date
Authors
Advisor
Related Publication Citation
DRUM DOI
Abstract
This paper focuses on determining how children with autism (ASD) approach word learning. Prior research has determined that non-autistic (NA) children exhibit word learning biases to assist with paring down potential or unlikely referent alternatives to effectively learn new word meanings. Some children with autism exhibit some of these word learning biases, but it is unclear if they exhibit two specific biases: the taxonomic and whole object biases. This project aims to investigate if children with ASD exhibit these biases, and if children with ASD perform similarly on the experimental tasks to their NA peers. 22 NA children ages 2-10 years participated in an analysis comparing their performance to that of children in previous studies and 11 ASD-NA pairs were matched for receptive language. Results suggest that children with ASD exhibit a taxonomic bias regardless of a label being presented or not, and that they exhibit a whole object bias. No statistically significant differences between groups arose for the taxonomic or whole object tasks, suggesting that children with ASD perform similarly to their NA peers. We anticipate this research will enhance our knowledge of how children with autism engage in the word learning process.
Notes
The proposed project included a brief play session, taxonomic bias task, whole object bias task. The play session was included to build rapport with the participant. The taxonomic task involved children selecting which object from a set of taxonomically similar, thematically similar, and unrelated objects they believed is most similar to a standard object upon hearing a novel label or no label condition. For instance, children might hear, “Can you point to the dax?” (label) or “Can you point to another kind of thing?” (no label). The whole object bias task involves children pointing to which object—from a set of a whole object and its two individual parts, side-by-side—they believe is attributed to the novel label (e.g. “Can you find the bif?”). Data analysis included a two-sample unpaired t-test to analyze the difference scores from the taxonomic bias task in children with ASD and NA children. For the whole object bias task, a two-sample unpaired t-test was utilized to compare the number of times children with ASD select the whole object choice compared to the number times NA children who make this same choice. Correlations were conducted between taxonomic and whole object task performance and distractibility, receptive language, and social communication, respectively.