Challenges of Adaptive Graphical User Interfaces

Adaptive graphical user interfaces (GUIs) automatically tai-lor the presentation of functionality to better fit an individual user’s tasks, usage patterns, and abilities. A familiar example of an adaptive interface is the Windows XP start menu, where a small set of applications from the “All Programs ” submenu is replicated in the top level of the “Start ” menu for easier access, saving users from navigating through multiple levels of the menu hierarchy (figure 1). The potential of adaptive interfaces to reduce visual search time, cognitive load, and motor move-ment is appealing, and when the adaptation is successful an adaptive interface can be faster and preferred in comparison to a nonadaptive counterpart (for example, Gajos et al. [2006], Greenberg and Witten [1985]). In practice, however, many challenges exist, and, thus far, evaluation results of adaptive interfaces have been mixed. For an adaptive interface to be successful, the benefits of correct adaptations must outweigh the costs, or usability side effects, of incorrect adaptations. Often, an adaptive mechanism designed to improve one aspect of the interaction, typically motor move-ment or visual search, inadvertently increases effort along another dimension, such as cognitive or perceptual load. The result is that many adaptive designs that were expected to con-fer a benefit along one of these dimensions have failed in prac-tice. For example, a menu that tracks how frequently each item is used and adaptively reorders itself so that items appear in order from most to least frequently accessed should improve motor performance, but in reality this design can slow users down and reduce satisfaction because of the constantly chang-ing layout (Mitchell and Schneiderman [1989]; for example, fig-ure 2b). Commonly cited issues with adaptive interfaces include the lack of control the user has over the adaptive process and the difficulty that users may have in predicting what the sys-tem’s response will be to a user action (Höök 2000). User evaluation of adaptive GUIs is more complex than eval