The mystery of how instruction leads to learning of disciplinary-based knowledge is examined through a fine grained analysis of a learning episode taken from a clinical interview in physics. The analysis employs two distinct theoretical perspectives, the Knowledge in Pieces perspective on conceptual change (diSessa, 1993) , and a distributed perspective of learning (Stevens & Hall, 1998) , linking the student's knowledge in-use and indevelopment to the interaction with the interviewer/tutor and the learning artifacts.