Work

Evaluating user performance in nonvisual bullseye menus

Year: 1997
Type: dissertation
Abstract: Standard GUIs today rely heavily on visual feedback. Nonvisual interfaces exist, but are far less common. Many human performance models which involve visual interfaces have been developed, such as Fitts' Law, but far fewer performance models have been developed for nonvisual interfaces. This thesis explores some possible models of a nonvisual bullseye menu system. A bullseye menu is a series of concentric circles divided into sectors. We call the region between two neighbouring concentric circles a ring. The category of choice is determined by the direction in which a user moves the stylus or mouse from a relative origin, and menu item selection is determined by the distance the user moves the stylus from the origin. If a nonvisual signal is emitted whenever the stylus is rolled over a circle, then the menu can be implemented nonvisually. We implemented the interface first with auditory feedback, using a simple beep, and then with tactile feedback, using a mechanism we created for this purpose. This thesis proposes models for performance time in nonvisual bullseye menus. We performed an experiment to test these models. Results showed Fitts' Law predicted time to target less accurately than a model where performance time is a linear function of target's distance. The model was extended to include a representation of the dispersion of the results, in order to provide a more complete description of the data. We ran an analysis of variance on the data collected to see whether performance time was influenced by feedback mechanism, ring width, target direction and target ring. Finally, some qualitative results based on user preferences are given.
Language: English
Open Access status: green
OpenAlex ID: W3177365631
Cites: 21
Cited by:
Related to: 10