Abstract
Recently, assistive devices designed for computer systems have proliferated, including those designed to support users with visual impairments (i.e., screen-readers). However, research examining how individuals use screen-readers to access the Internet is virtually non-existent. Therefore, the purpose of the reported study was to examine the behavior of a user with visual impairment, via protocol analysis, while using the World Wide Web to find information. We classified her behavior into an HCI-relevant model, i.e., Norman's (1988) seven stages of action, in order to identify potential usability bottlenecks. The results indicated that executing actions and, more notably, interpreting the system state were the most frequent and time-consuming tasks. In addition, the results suggested that the user had difficulty determining the effects of her control inputs on system status, as well as determining whether or not goal-relevant information was present on the current page. For screen-reader design, this suggests that there are possible usability problems in interfacing the user with the screen-reading software and the way textual information is aurally displayed to the user.
Get full access to this article
View all access options for this article.