Abstract: | Self-perception processes have been postulated to occur only to the extent that initial attitudes are weak. The present research asked whether the outcome of such a process is a strengthening of the attitude in question. Two experiments investigated the accessibility of attitudes from memory following self-inference from behavior. Experiment 1 examined the consequence for attitude accessibility of reviewing and considering previously performed religious behaviors that were recent and primarily unmanded versus distant in time and "manded" in nature. Experiment 2 involved the performance of a new behavior that was either required or freely chosen. In each case, control subjects either did not review prior behaviors or did not perform a new behavior. In both experiments, attitude accessibility, as measured by the latency of response to attitudinal inquiries, was enhanced by the consideration or performance of unmanded behavior, but not by manded behavior. The relevance of this finding to issues regarding attitude-behavior consistency and attitudinal persistence is discussed. |