Modern psychology judges its progress and products by a variety of criteria. Reviewing a number of paradigms in current cognitive psychology, Claxton (1988) suggested that the research community gives no less than thirteen answers to the question: “How do you tell a good cognitive theory when you see one?” (pp. 1-31) Each of the 13 criteria he mentions (e.g., experimental, computational, evolutionary) has enough adherents so that research programs are judged successful, even if their products meet perhaps only 1 of these standards of merit. Research activity in current cognitive science thus resembles a massively parallel search, in which most of Claxton’s 13 criteria for scientific success are suspended on any one search path so that individual research efforts can proceed unencumbered by a diverse set of otherwise paralyzing constraints. For example, in certain paradigms computational realization is the primary concern, mathematical formalization the major constraint in others, and in still others a necessary condition for a theoretical model may be a demonstration that the proposed cognitive mechanisms and processes could have emerged through human development or evolution. The eventual success of this divide-and-conquer venture, of course, hinges not so much on whether each of the many paradigms meets its own goals, but rather on whether we are somehow able to integrate the resulting array of research products into 69useful and coherent theory.