Williams, P, Yeshurun, Y, & Carrasco, M (2001, May). Masked or not, covert attention enhances spatial resolution: Support for signal enhancement. Vision ScienceS Society, Sarasota, FL. 79-B12.

Abstract
Introduction. Several authors have attributed covert attentional effects to external noice reduction and have questioned the existence of signal enhancement. Previously, Yeshurun and Carrasco (1999) showed that attention improved spatial resolution in acuity tasks, even when eliminating all the known sources of external noise -e.g., distracters and global masks. We attributed this finding to signal enhancement. Recently, however, Smith (2000) proposed that a local post-mask, which we had used, could be another source of external noise.

Goal: Using a supra-threshold Landolt square stimulus, we investigated whether transient covert attention enhanced the signal even in the absence of a local post-mask. Furthermore, we compared performance when a 'peripheral' cue preceded the target location with performance under two neutral cues, one 'central' and the other 'spread'.

Methods. In a 2AFC discrimination task, observers reported whether the right or left side of a Landolt square contained a gap. The target appeared alone at varying eccentricities along the horizontal and vertical meridians. In half of the blocks a local mask followed the target. Stimulus duration was adjusted for eacfh observer to attain about 80% overall correct performance. A precue either indicated both location and target onset (peripheral cue) or just target onset (neutral cue). The central cue appeared in the center of the display; the spread cue was composed of 4 bars each appearing in the center of a quadrant. The short duration of the trial precluded eye movements.

Results. Regardless of mask presence, performance was more accurate and faster in the peripheral cue than in either neutral cue condition, for all gap sizes and at all eccentricities. The cueing effect become more pronounced at farther eccentricities.

Conclusion. Having eliminated all possible sources of external noise, the attentional benefit, reflected in improved spatial resolution, must be attributed to signal enhancement.