neth.de > | <<


Forschung
Research

According to Kirsh (1995; Kirsh & Maglio, 1994) not all actions carried out during problem solving are meant to solve the problem. In this talk, we will examine two different kinds of 'non-pragmatic actions' in two different task domains: In an experiment on the execution of a routine skill (addition of coin values) we examine participants' spontaneous usage of complementary actions, which modify the problem to facilitate subsequent operations. It is argued that it is the result of complementary actions, rather than acting itself, which is responsible for the observed benefits in performance. In a second experiment we are trying to disentangle the notions of 'pragmatic' and 'epistemic' actions in a more traditional problem solving context. In classic studies of problem solving (e.g., Anzai & Simon, 1979) incidental learning ('learning by doing') appears to be a by-product of problem solving activity. However, it is conceivable that some actions are actually performed to learn more about the problem rather than to move closer to its goal state. Using a transformation task (Towers of Hanoi) we are trying to induce and isolate epistemic actions (by instruction or device support) and assess their consequences on problem solving performance and learning.

Please note: The experiments reported in this talk have been carried out in collaboration with Stephen J. Payne . Our joint research on epistemic actions has been inspired by the work of David Kirsh and Paul Maglio . For additional information and some fascinating ideas, please refer to their publications.