Latent inhibition refers to retardation of the development of a conditioned response when the conditioned stimulus (CS) is preexposed alone prior to its pairings with an unconditioned stimulus. Experiment 1 demonstrated this effect for rats trained in an appetitive conditioning procedure and confirmed that the effect is found when the target stimulus is presented in compound with another or with a range of other stimuli during preexposure. Previous work has shown that a latent inhibitor does not reliably reduce the level of conditioned responding supported by an excitatory CS when the 2 stimuli are presented in compound (in a summation test). In Experiments 2, 3, and 4 we demonstrate that preexposure in which the target stimulus is presented in compound with a novel event on every trial will render that stimulus effective in a summation test. This outcome is uniquely predicted by the account of latent inhibition proposed by Hall and Rodríguez (2010), which suggests that the latent inhibition effect is a consequence both of a reduction in the associability of the stimulus and of a process of inhibitory associative learning that opposes the initial expectation that a novel event will be followed by some consequence.