This study compared interaction with a computer vs. observation as learning situations for low and high ability student's learning of spatial visualization and geometric transformations. Thirty-two fifth grade boys took the Differential Aptitude Test, Space Relations Subset (DAT), and then participated in the experiment. Pre-test and post-test were static spatial visualization problems using polyominos. During treatment, subjects were paired, one interactively solving computer-based polyomino puzzles, the other observing a yoked monitor in a separate room. Think-aloud protocol was used throughout. Overall, no significant differences were found between the two conditions. However, high achievers on the DAT benefited significantly more from observing (p.5). Low achievers benefited marginally more from the interactive condition, showing an increased incidence of holistic mental rotation strategy following the interactive condition. (Contains 3 tables and 5 figures.)