Abstract
When a subject repeatedly views the same stimulus, the visual system exhibits a decreased response. This decrease in activity has been variously referred to as “adaptation,” “repetition suppression” and “priming,” and has been observed using numerous techniques including single-cell recordings in primates and fMRI in humans. Previous studies examining adaptation in humans have simultaneously varied several parameters which may exert diverging effects. Here we examined the separate role of three parameters on adaptation and behavioral priming using an event-related fMRI design. The parameters that were manipulated were: the number of stimulus repetitions (n:1–8), the time between repetitions (ISI:0–16 or more sec), and the number of intervening stimuli between repetitions (i:0–8 or more). Subjects were instructed to classify animal images. Reaction time and accuracy were recorded during the fMR scans. Regions of interest involved in object recognition consistently exhibited repetition suppression. The dependence on n was graduated, adapting steadily with the first 4–5 repetitions of an image. Adaptation was significant for all values of i and ISI and maximal when there were no intervening stimuli. Surprisingly, increasing the number of intervening stimuli did not much change the amount of adaptation. Subject reaction-time data exhibited a similar but not identical dependence on these parameters. These results indicate that adaptation in the visual system is not a short-term phenomenon, but can persist with little reduction over many intervening images and long periods. Ongoing work is aimed at further elucidating the relationship between behavioral priming and fMR adaptation.