Modeling Users with Disabilities Interacting with Computers Through Assistive Technology
Speaker
Rich Simpson
Department of Rehabilitation Science and Technology, University of Pittsburgh
When
-
Where
Newell-Simon Hall 1305 (Michael Mauldin Auditorium)
Description
Optimizing the fit between technology and the user is critical for realizing the potential benefits of assistive technology, but clinicians (who are responsible for identifying and configuring assistive technologies for their clients) are frequently presented with the daunting task of identifying the most appropriate device and configuration for a client based on one or two hours of interaction. Such a short period of time makes it extremely difficult to empirically compare a wide range of devices and configurations, and impossible to determine how skilled the client will become on any one device/configuration with practice. The device a clinician chooses for a client will often not be replaced for several years, and the configuration used in the first device is often carried over to subsequent devices, so this decision is crucial.
One potential solution to this problem is an accurate mechanism for modeling the performance of individuals with disabilities when using computer access and AAC devices. An accurate modeling technique would help a clinician to identify potential devices and device configurations that are most likely to be successful for a client. Existing user modeling methods have already proven their usefulness in the product design and evaluation process. However, most modeling methods have been designed for, and verified with, able-bodied users. While these models might provide useful qualitative predictions of task performance and learning time for individuals with disabilities, their ability to provide accurate quantitative predictions is limited.
Investigators at the University of Pittsburgh are exploring user modeling techniques that provides accurate quantitative predictions of performance, learning time and fatigue for individuals with disabilities. Our ultimate goal is to develop a software tool that allows clinicians to incorporate the user modeling technique into their existing clinical assessment processes.
Speaker's Bio
Richard Simpson, PhD, ATP, received a BS in Computer Science from Virginia Tech in 1992. At the University of Michigan he earned an MS in Bioengineering in 1994, an MS in Computer Science and Engineering in 1995, and a PhD in Bioengineering in 1997. Dr. Simpson was certified as an Assistive Technology Practitioner in 1997. He is an assistant professor at the University of Pittsburgh, in the Department of Rehabilitation Science and Technology. His research interests include human-computer interaction, alternative computer access technology, augmentative communication and rehabilitation robotics.
Speaker's Website
http://www.shrs.pitt.edu/ris20/
Host
Brad Myers