Clearly, there are benefits for knowing what someone will do with certain knowledge. For instance, if one knows that a process control operator's model of a system is different (in some ways) from a troubleshooter's model of a system, training regimens can be designed to accommodate and support their respective tasks. Knowledge about information deemed important by users can also be directed toward effective training sessions and user guides.
If our theoretical assumptions are supported, early user-system interaction should be characterized by the user's relying on superficial features of the system. Perhaps the user will spend most of his or her time attempting to recognize and operate these features of the new system. With some experience, however (and this can be simulated), the user should no longer be concentrating on these surface features. In fact, the superficial features become irrelevant during interaction with the system. The interface should provide more than one interaction "mode" so that important relational features are now easily accessible and clearly displayed. The number of adapting modes is an empirical question and could also be simulated using the theory. Finally, knowing the kinds of features important in user models of a system could help the user interface designer in providing useful associations and analogies to the user. As Halasz and Moran ( 1982) argue, providing the user with a conceptual model can be productive for learning. In order to determine which associations and analogies might be deemed most useful, theoretical simulations could be run and then matched with usability data later.
Knowing the kinds of models people have or will generate gives system designers and human factors practitioners the leverage to design better systems for those models. Knowing the level of expertise a particular user has will be important as more intelligent user interfaces are designed, for example, for expert systems. A quantitative theory of mental model acquisition and maintenance, as discussed above, could prove beneficial in designing flexible interface communication that adapts to the user as his or her interactions with the system give way to different levels of system understanding.
Anderson J. R. ( 1987). "Skill acquisition: Compilation of weak-method problem solutions". Psychological Review, 94, 192-210.
Ashby W. R. ( 1956). Introduction to cybernetics. London: Chapman and Hall.
Bainbridge L. ( 1979). "Verbal reports as evidence of the process operator's knowledge". International Journal of Man-Machine Studies, 11, 411-436.
Bainbridge L. ( 1981). "Mathematical equations or processing routines". In J. Rasmussen & W. B. Rouse (Eds.), Human detection and diagnosis of system failures (pp. 259-286). New York: Plenum.
Bainbridge L. ( 1986). "Asking questions and accessing knowledge". Future Computing Systems, 1, 143-149.
Barnett B. ( 1989). Information processing components and structural knowledge representations in pilots' judgments. Unpublished doctoral dis-