A reevaluation of Rescorla's early dictums about Pavlovian conditioned inhibition.