Scott's research interests center around fundamental limits on what can efficiently be computed in the physical world. This has entailed studying quantum computing, the most powerful model of computation we have based on known physical theory. Scott's work on this subject has included limitations of quantum algorithms in the black-box model; the learnability of quantum states; quantum proofs and advice; the power of postselected quantum computing and quantum computing with closed timelike curves; and linear-optical quantum computing. Scott also maintains an active interest in many topics in classical theoretical computer science, including circuit lower bounds, computational learning theory, and communication complexity.