Computation as a Tool for Discovery in Physics: A Personal History Barry I. Schneider Program Director for Theoretical Physics Physics Division National Science Foundation Computation has now taken its place alongside theory and experiment as a third approach to providing insights into the behavior of physical systems. For many problems, computation is the sole approach available to achieve meaningful results. Since the early 1970’s my own research has been directed toward developing and applying the most effective theoretical techniques to a number of problems in computational atomic and molecular physics. I will draw from this research on electron-molecule scattering and ultracold, trapped atomic gases to illustrate what can be learned from computation. The talk will conclude with some general comments concerning how the field of computational physics is progressing.