Three training methods were compared in a computer text editing situation. Training varied on the degree of constraint imposed on the behavior of learners. For complex methods, completely constrained and completely unconstrained training situations led to worse test perfomance than an unconstrianed training condition in which subjects were informed about their compliance with a method and allowed to reset the system at will. The results argue against a strict “training wheels” approach to learning environments but support the general notion of “guided exploration,” especially for complex methods.
Get full access to this article
View all access options for this article.
References
1.
AndersonJ.R. (1983). The architecture of cognition.Cambridge, MA: Harvard University Press.
2.
CardS.MoranT.NewellA. (1983). The psychology of human-computer interaction.Lawrence Erlbaum Assoc.
