The contingency between environmental events and behavior has proven to be a useful concept in the study of both behavior and cognition. There is common ground in the definition of contingency in both domains, but interpretations of the basis of its action differ. For behavior analysts the contingency acts through both its direct, response-strengthening effect and indirectly through its function as a discriminative stimulus. Cognitive accounts, as represented in the work of both Bower and Watson, focus more on the organism's detection and interpretation of the contingency as the basis of its action. Despite such conceptual differences, Watson's quantitative descriptions of contingency effects seem relevant to feedback functions that describe reinforcement schedule performance and, as such, may bear on research involving combinations of response-dependent and response-independent food presentations and on superstitious behavior.