Abstract Philip Brickman 1978, in New directions in attribution research (pp. 5–34, Vol. 2, Hillsdale, NJ: Erlbaum) proposed that an actor's subjective sense that an experience is “real” (vs “unreal”) depends on two factors: (1) the degree of internal correspondence, i.e., the degree to which the actor's behavior corresponds to his or her feelings and (2) the degree of external correspondence, i.e., the degree to which the actor's behavior correpsponds to external consequences. Prior to his death, Brickman developed a brief measure, the Internal-External Correspondence Scale (ICS), designed to assess individual differences in the preference that one's behavior be internally versus externally correspondent. The present authors have refined Brickman's original, unpublished version of the ICS and report evidence for the internal consistency, test-retest reliability, and construct validity (e.g., convergent/discriminant validity and predictive validity) of the revised version. This evidence supports specific predictions found in Brickman's (1978) theoretical formulation. It further reveals that internal correspondence functions as a personal moderating variable, such that the behavior of high internally correspondent subjects is strongly correlated with their feelings and dispositions, whereas the behavior of low internally correspondent subjects is only weakly correlated with their feelings and dispositions.