After earning my doctorate, I worked as a research associate with Kurt VanLehn. We worked together in the Department of Psychology at Carnegie Mellon University for one year. I then moved with him to the Learning Research and Development Center at the University of Pittsburgh, where I worked for two years. I participated primarily in two research projects, developing computer models to explain specific psychological phenomena in learning and problem solving. One system models how children invent new strategies when learning to add. The other models the use of explanation and analogy by university students learning classical physics. The theme tying this research together involves the interaction of learning mechanisms with existing knowledge to produce predictable changes in behavior. Understanding these interactions should greatly aid our ability to educate/train/write textbooks as well as to study the sources of differences in behavior between individuals.