Usability & Likability of the Virtual Environment for Social Information Processing (VESIP TM) for Children with and without Autism Spectrum Disorders

Background:  

There is a well-articulated theoretical model for understanding social information processing (SIP; Crick & Dodge 1994, 1996). Most SIP assessments are vignette-based interviews that require significant training to administer and score. It is unclear how well interviews about social problems approximate the experience of being in the midst of a social problem. To advance science and practice, we are developing the Virtual Environment for Social Information Processing (VESIP^TM), a computerized simulation in which children adopt the role of an avatar and use a game controller to navigate challenging social situations and engage in real-time social decision-making. 

Objectives:  

First, we tested the usability and feasibility of our prototype in typically-developing (TD) children and children with autism spectrum disorders (ASD). Our second objective was to show initial evidence that the VESIP^TM is sensitive to diagnostic differences. Our overarching goal is to create a user-friendly, scientifically sound, and automated system for SIP skill assessment.

Methods:  

20 children (10=ASD, 10=TD; ages 8-14 years) participated. ASD diagnoses were confirmed through parent interview and administration of the ADOS.

During the VESIP^TM, the child assumed the role of a character and navigated the social situation in an animated scene using a game controller. The prototype ambiguous provocation situation involved a child being bumped by another child. Using the controller, children selected (1) an emotional response, (2) actions to engage at first insult versus after thinking about it, (3) how well one is able to carrying out solutions, and (4) intent of the provocateur. As selections were made, avatars acted accordingly. Afterwards, children rated the usability and likability of the VESIP^TM, reported general impressions (e.g., likes and dislikes), and described the tool in their own words (demonstrating task comprehension). Children also completed some interview questions that paralleled key decision points in VESIP^TM.

Results:  

On a scale of 1-5 (1=very easy, 5=very hard), children with and without ASD rated the the instructions as being very easy to understand (ASD mean=1.2, TD mean=1.7) and the tool as being very easy to use (ASD mean=1.3, TD mean=1.5). Again on a scale of 1-5 (1=very fun, 5=no fun), children rated the tool as being moderately enjoyable (both groups, mean=3). Eighty-seven percent of responses on the computerized assessment matched responses to the same interview questions. Group differences were most evident in the following ways: (1) 50% of TD children chose pro-social and competent goals, compared to no children with ASD, (2) children with ASD were less likely to elect to have an adult intervene; and (3) children with ASD tended to report less confidence in their ability to carry out a solution to the problem.

Conclusions:  

This pilot study data supports the usability and likeability of the VESIP^TM and its sensitivity to diagnostic differences. We are further developing the tool by incorporating child feedback and adding other challenging social situations. We expect this tool to (1) allow for more efficient SIP assessment, (2) have greater ecological validity over other tools, and (3) identify specific SIP steps for remediation with targeted interventions.