Imagery in Science Learning in Students and Experts

Abstract

In this chapter we review three studies done at the University of Massachusetts which have implications for the role of visualizations in learning. We then identify common themes in these studies and their implications for instructional design. We describe two studies of model-based and imagerybased strategies for teaching conceptual understanding in science. To investigate these issues we have conducted learning studies where students are asked to think aloud while learning from innovative science lessons. The goal of these studies is to describe learning processes and teaching strategies that lead to the construction of visualizable models in science. A third study examines similar processes in scientists thinking aloud about an explanation problem. The studies develop observable indicators for the presence of imagery and use them to support a theory of how imagery is being used. A common theme across the studies is that dynamic imagery from one context could be transferred to a new model being constructed for a new context. In some cases new models can be grounded on runnable prior knowledge schemas (e.g. physical intuitions) that can generate dynamic imagery. We suspect that developing students' runnable mental models in this way provides a deeper level of conceptual understanding. Describing expert novice similarities in terms of hidden nonformal reasoning processes and the use of intuitive knowledge structures allows us to build a map of the conceptual terrain in students that should have important implications for instructional design.