Effects of Systematic Social Skill Training on the Social-Communication Behaviors of Young Children with Autism During Play Activities

A systematic social skills training intervention to teach reciprocal sharing was designed and implemented with triads of preschool-age children, including one child with an autism spectrum disorder (ASD) and two untrained classroom peers who had no delays or disabilities. A multiple-baseline research design was used to evaluate effects of the social skills training intervention on social-communication and sharing behaviors exhibited by the participants with ASD during interactive play activities with peers. Social-communication behaviors measured included contact and distal gestures, touching peers and speaking. Four sharing behaviors were also measured, including sharing toys and objects, receiving toys and objects, asking others to share, and giving requested items. Results indicated considerable gains in overall social-communication behaviors. The greatest improvements were observed in the participants’ use of contact gestures and speaking. Slightly increasing trends were noted and suggested that participants with ASD made modest gains in learning the sharing skills taught during social skills training lessons. Social validity data indicate that participants with ASD and peer participants found the intervention appropriate and acceptable, and staff perception ratings indicated significant changes in the social skills of participants with ASD. Study outcomes have practical implications for educational practitioners related to enhancing social-communication and social interactions of young children with ASD. Study limitations and future directions for research are discussed.

Layered Clustering Communication Protocol for Wireless Sensor Networks

In this paper, we propose a Layered Clustering Hierarchy (LCH) communication protocol for Wireless Sensor Networks (WSNs). The design of LCH has two goals: scalability and energy-efficiency. In LCH, the sensor nodes are organized as a layered clustering structure. Each layer runs a distributed clustering protocol. By randomizing the rotation of cluster heads in each layer, the energy load is distributed evenly across sensors in the network. Our simulations show that LCH is effective in densely deployed sensor networks. On average, 70% of live sensor nodes are involved directly in the clustering communication hierarchy. Moreover, the simulations also show that the energy load and dead nodes are distributed evenly over the network. As studies prove that the performance of LCH depends mainly on the distributed clustering protocol, the location of cluster heads and cluster size are two critical factors in the design of LCH.