Examining the Relationship between Behavioral Repetition Priming and fMRI Repetition Suppression

Abstract

Priming refers to a change in the ability to identify, produce, or classify a stimulus as a result of a previous encounter with the same or a related stimulus. Recent neuroimaging studies often found behavioral priming to co-occur with a reduction in neural activations in various cortical regions, which is called repetition suppression. It is thought that repetition suppression is closely related to behavioral priming, and may even be the underlying neural mechanism that supports priming. However, current literature still has several unsolved questions about the relationship between repetition suppression and priming. The present dissertation set out to further elucidate their relationship. In Study 1, a mirror-word identification task was used to limit overlap between study and test to a primarily perceptual level with little or no conceptual overlap nor top-down modulation. Repetition suppression was found in visual perceptual and frontal phonological regions involved at both study and test, supporting a "component process" view that repetition suppression and priming can occur at a perceptual level with limited conceptual or top-down processes involved. In Study 2, three perceptual priming tasks and one conceptual priming task were used to directly examine component process view's prediction that perceptual priming would be correlated with posterior repetition suppression and conceptual priming would be correlated with frontal repetition suppression. The results showed that both perceptual and conceptual priming involved repetition suppression in both frontal and posterior perceptual regions, at least when measured with our paradigm and tasks, and both frontal and posterior repetition suppression effects were correlated with behavioral priming in all four perceptual and conceptual priming tasks. This finding suggests that both frontal and posterior perceptual regions are involved in perceptual and conceptual priming, and that they are most likely working in concert with one another during priming, as exemplified by an interactive view of priming. Taken together, our data suggest that priming may be supported by several different underlying mechanisms, such as bottom-up processes (component process view of priming), top-down modulation and frontal-posterior interaction/synchrony.