Abstract
Little is known about the role of extra-foveal processing in VSTM, despite its potential importance in maintaining active scene memory. This study investigated extra-foveal processing in VSTM in two object-position recognition experiments using five observers. Unfamiliar, nonverbal 1/f noise discs served as stimuli, minimizing confounds arising from verbal labeling and semantic association. At study, observers viewed 5 equally spaced 2 deg stimuli in a 13 deg ring, either simultaneously (experiment 1) or sequentially (experiment 2). A random start position was pre-cued with a spatial probe. In experiment 1 (extra-foveal processing enabled), stimuli were viewed consecutively in a clockwise direction, guided by an auditory prompt every 650ms until all study stimuli were fixated. In experiment 2 (extra-foveal processing disabled), the auditory prompt coincided with the onset of the next stimulus and the offset of the previous stimulus. At test, following a 1000ms ISI, a centered target stimulus was displayed, flanked by spatial markers at the 5 study stimulus positions. Observers attempted to identify the spatial position occupied by the target in the preceding study display (5AFC). Hit rates were calculated for each temporal index (1=earliest, 5=latest) at which target stimuli were presented in the study display. In general (over the earliest 3 temporal indices), hit rate was significantly higher in experiment 1 than in experiment 2 [F(1,4)=26.18, p[[lt]]0.01]. At the fourth temporal index, hit rate was not significantly different between the experiments [t(4)=0.49, p=0.65]. At the fifth temporal index, hit rate was significantly lower in experiment 1 than in experiment 2 [t(4)=−2.75, p=0.05]. Results suggest extra-foveal processing in VSTM for object-position recognition exhibits a trade-off between facilitation and inhibition. Facilitation is incurred after the target has been foveated, decreasing the impact of retro-active interference; conversely, inhibition is incurred before the target has been foveated, increasing the impact of pro-active interference.