Computational activities in Matter & Interactions, an introductory calculus-based physics course, provide students with  experience applying a small set of fundamental principles to model a wide range of physical systems.

This study explores how students complete interpretation and prediction tasks when presented with functional but incomplete VPython programs. Rather than asking students to modify programs, students interpret programs missing key lines of code corresponding to the algebraic form of physics principles, and draw predictions of how the simulation would evolve. They then run the program to evaluate the prediction. This study specifically examines how participants: 1. use physics while interpreting the program code and creating a prediction; 2. evaluate their understanding of the program and goals at the beginning of a modification task.

Study participants working in groups were recorded completing three activities. Video data analysis showed participants had little difficulty interpreting physics quantities, generating predictions, or determining how to modify incomplete programs. When trying to predict the motion of the objects in the simulation, many turned to their knowledge of how the system would evolve if it represented a real-world physical system. Participants rarely interpreted lines of code in the computational loop during the first activity, but most used their physics knowledge to interpret computational loops during latter activities.

Computational activities in the Matter & Interactions curriculum were revised based on these findings. The modified activities ask students to create an additional whiteboard prediction for the time-evolution of the real-world phenomena which example programs will eventually model. This thesis shows how comprehension tasks effective in improving reading comprehension (Palinscar and Brown 1984) also help students apply their physics knowledge in interpreting and understanding computational models.