Date of Award

Summer 2007

Document Type


Degree Name

Doctor of Philosophy (PhD)


Biomedical Engineering

First Advisor

Stanley Cronk


Persons with a severe disability often use scanning as an indirect selection technique for operating augmentative and alternative communication aids and computer access. For information that can be organized in advance, including lists of communication elements such as words and phrases, users often employ rate enhancing scanning methods like the row-column scanning technique. However, row-column scanning requires selection elements to be grouped into defined rows and columns, and therefore does not work well with Internet browsing due to the non-grouped layout of HTML pages.

This work attempts to develop an improved scanning technique for Internet browsing by designing interfaces to compare two contemporary scanning techniques with the overscan scanning technique, also known as the critically damped selection technique. The hypothesis of this investigation is that the overscan technique is a viable technique for persons with a severe physical disability to use to access the Internet. Alphabetic and Internet browsing interfaces were designed to test the error rates, throughput, key press times, reaction times, and activation forces for three different scanning methods: linear, row-column, and overscan. The effectiveness of the interface was determined by testing each interface with individuals without a disability, and a Goals, Operators, Methods, and Selection Rules (GOMS) model for the three scanning methods tests was developed.

The throughput of the overscan technique was a significant improvement over the linear scan technique for both the alphabetic and Internet interfaces. The individuals testing the interface were able to realize this increased throughput while maintaining error rates which were slightly less than the error rates measured while using the row-column interface. The error rates for the overscan and row-column scanning techniques were greater than the error rates for the linear scanning technique, but the time lost on erroneous selections was much less than the time gained through the use of the overscan and row-column selection techniques.

The overscan technique was shown to be a viable scanning technique for Internet browsing. Use of overscan as a method of indirect selection for Internet browsing could connect individuals to the Internet who are not now linked to this electronic communication medium.