This research project builds upon a recently completed and related project undertaken as part of the U.S. Department of Education's SBIR program. It directs attention to cognitive and adaptive difficulties which accompany attempts by developmentally disabled individuals to use contemporary servo-electronic devices such as robots, computers, automatic wheelchairs and similar devices. Based on an analysis of those difficulties, the proposal advances a unique neural network adaptation to overcome these inherently frustrating problems of adapting these systems to a developmentally disabled population. The ready availability and low cost of recently introduced PC-based network boards provides a readily available means to examine the key tenants attendant with developing a system that compensates for physical limitations of the developmentally disabled. Specific project activities are examining various elements in the state- of-the-art and undertaking preliminary studies on network paradigms that might be efficacious in developmentally disabled individuals. Recently completed research and the proposal demonstrate the significant benefits which are attendant with success in this development. Specific issues related to true independence of handicapped individuals are described along with the inherent benefits of the system. Neural networks exhibit compensatory capabilities which are unique. When integrated with an overall plan to develop the cognitive and affective quality of life experienced by the developmentally disabled, it is believed systems configured around this modality will provide the first low cost devices which enable the disabled to live a truly independent life among their peers.
