ABSTRACT

This chapter reviews computational representations of human behavior involving three training principles discussed in preceding chapters (especially chapters 2, 3, and 5): Speed–accuracy trade-off attributable to fatigue, training difficulty, and stimulus-response compatibility. Effects of these three training principles were modeled using the ACT-R cognitive architecture (Anderson & Lebiere, 1998) and the instance-based learning (IBL) theory (Gonzalez, Lerch, & Lebiere, 2003). The use of similar memory principles in all three projects resulted in the implementation of an IBL tool (Dutt & Gonzalez, 2011), which provides a computational framework that facilities building computational models using ACT-R and IBL theory. The last section of this chapter summarizes the IBL tool and concludes with the benefits of using computational representations of learning and training principles: to develop an understanding of the learning process in a variety of tasks; to predict learning effects from training principles; and most importantly, to demonstrate the generality of computational principles and representations from the ACT-R architecture and IBL theory.