Broadening Accessibility Through Special Interests: A New Approach for Software Customization

Robert R. Morris, Kirschbaum C., Rosalind W. Picard


Individuals diagnosed with autism spectrum disorder (ASD) often fixate on narrow, restricted interests. These interests can be highly motivating, but they can also create attentional myopia, preventing individuals from pursuing a broad range of activities. Interestingly, researchers have found that preferred interests can be used to help individuals with ASD branch out and participate in educational, therapeutic, or social situations they might otherwise shun. When interventions are modified, such that an individual’s interest is properly represented, task adherence and performance can increase. While this strategy has seen success in the research literature, it is difficult to implement on a large scale and therefore has not been widely adopted. This paper describes a software approach designed to solve this problem. The approach facilitates customization, allowing users to easily embed images of almost any special interest into computer-based interventions. Specifically, we describe an algorithm that will: (1) retrieve any image from the Google image database; (2) strip it of its background; and (3) embed it seamlessly into Flash-based computer programs. To evaluate our algorithm, we employed it in a naturalistic setting with eleven individuals (nine diagnosed with ASD and two diagnosed with other developmental disorders). We also tested its ability to retrieve and process examples of preferred interests previously reported in the ASD literature. The results indicate that our method was an easy and efficient way for users to customize our software programs. While we believe this model is uniquely suited for individuals with ASD, we also foresee this approach being useful for anyone that might like a quick and simple way to personalize software programs.

Related Content