Implicit attitude formation through classical conditioning

We sought to demonstrate that attitudes can develop through implicit covari ation detection in a new classical conditioning paradigm. In two experiment s purportedly about surveillance and vigilance, participants viewed several hundred randomly presented words and images interspersed with critical pai rings of valenced unconditioned stimuli (USs) with novel conditioned stimul i (CSs). Attitudes toward the novel objects were influenced by the paired U Ss: In a surprise evaluation task, the CS paired with positive items was ev aluated more positively than the CS paired with negative items. This attitu dinal conditioning effect was found using both an explicit measure (Experim ents I and 2) and an implicit measure (Experiment 2). La a covariation esti mation task involving the stimuli presented in the conditioning procedure, participants displayed no explicit memory for the pairings.