ABSTRACT
Geometrical concepts involve an intimate mixture of perception and cognition. Most geometry problems are either posed in the form of a diagram, or require construction of a diagram for a successful solution. Yet, unlike the case for reading and arithmetic, a serial algorithm for scanning or constructing diagrams is not taught to students of geometry. Specifically, when learning to read, students are taught to proceed from word to word, from left to right. When learning column arithmetic, they are taught to proceed from right to left through the columns, and from top to bottom within each column. When taught geometry, on the other hand, students are not taught how to look at the diagrams — they must learn an efficient eye movement pattern on their own. Empirical studies of eye movement patterns during reading and arithmetic show that individuals performing these tasks successfully indeed follow an orderly eye movement pattern with very few deviations (Epelboim, Booth & Steinman, 1994; Suppes, 1990). The nature of eye movements performed in order to solve geometry problems is not known. These eye movements and what they can tell us about the underlying mental operations are the topic of our study.
