Abstract
Introduction. Only two studies have explicitly manipulated and shown the effect of exogenous feature-based attention (FBA; Lin, Hubert-Wallander, Murray, & Boynton, 2011; Qian & Liu, 2015), both reporting effects on reaction time (RT) that emerge faster than endogenous FBA. Endogenous FBA is known to spread across space but for exogenous FBA, its distribution across space and effect on accuracy are unknown. Here, we investigated the spatial distribution of exogenous FBA. Methods. Participants discriminated the orientation (vertical or horizontal) of one colored ellipse within an array of five colored circles. Each stimulus was a different color, and could appear at any of eight isoeccentric locations, at all cardinal (horizontal and vertical) and intercardinal locations. An uninformative pre-cue was presented at the center of the screen, the color of which matched either the target (valid) or one of the distractors (invalid). Three ISIs between the cue and the stimulus display were used in separate blocks – 100, 200, and 300 ms. Results. There was a significant interaction between ISI, Feature-Cue Validity, and Location. For locations along the cardinal axes (horizontal and vertical), accuracy was higher for valid trials compared to invalid trials. This effect emerged only for the short, 100ms ISI. No differences in accuracy were observed between valid and invalid trials for intercardinal locations. Further, there was no significant effect of RT, indicating no speed-accuracy tradeoff. Conclusions. Our results suggest that exogenous FBA is not evenly distributed across space. Specifically, exogenous FBA facilitates performance as fast as 100ms after a cue appears, and this effect is constrained by the location of the stimulus, emerging only along the cardinal axes, and not at the intercardinal locations. This contrasts with endogenous FBA, which spreads across space and takes about 500 ms to be deployed.
Meeting abstract presented at VSS 2017