Understanding Expertise in Physics

Steven Wolf is a Ph.D. candidate in the Department of Physics and Astronomy, studying physics teaching and learning. His aim is to guide students toward an expert-like understanding of physical phenomena. But what does it mean to be “expert-like?”

More than 30 years ago, a seminal experiment was conducted that first qualitatively described the difference in how experts and novices first approach a physics problem. However, the statistical side of this experiment has never been sufficiently and systematically explored. Wolf puts the analysis of this experiment into the more appropriate framework of graph theory. “We’ve already established that the expert-novice distinction is not the prominent discriminator between categorizations and that experimental success strongly depends on the composition of the problem set,” says Wolf. Finding the optimal set is a combinatorial problem, involving the calculation of the distance between two graphs at each step. This task is computationally intense, but maps well on a parallel architecture. “My work may well be the first application of supercomputing to education research, and the computational resources at iCER will allow me to graduate before I reach retirement age,” Wolf explains.