Abstract
In studies of the behavioral effects of attention, performance is often rendered difficult by degrading the stimulus in some way (brief stimulus presentations, addition of noise, reduction of contrast). The goal of this research was to determine which stimulus attributes are affected by voluntary spatial attention. Subjects performed orientation judgments on oriented Gabor patches, and spatial attention was directed using spatial-cueing paradigms. In the first experiment, an anti-cueing design was used (Posner, Cohen, & Rafal, 1982): a peripheral cue at one location indicated a higher probability of the target appearing at the opposite location. Better performance at the opposite location at long cue-to-stimulus intervals is the result of voluntary attention, whereas better performance at the cued location at short intervals indicates involuntary attention. Visual “white pixel noise” was added to the stimuli. Orientation discrimination thresholds were affected by voluntary but not by involuntary attention. In experiments 2–4, a direct predictive cue was used (Posner, 1980). These experiments determined the contributions of external noise and the size of the orientation discrimination on attention effects. Experiment 2 compared predictive and nonpredictive cues with no external noise in a fine orientation discrimination. Experiment 3 replicated this experiment with fiduciary markers (e.g., Gould et al., 2007) to ensure that there was no location uncertainty. In Experiment 4, the orientation difference was large, but performance was limited by adding pixel noise. We found significant effects of voluntary attention, both with fine orientation judgments with no external noise and also with coarse judgments and external noise. Finally, unlike the previous experiments, limiting performance by lowering contrast showed no effects of attention when the orientation discrimination was large. These results suggest that attention enhances both fine and coarse orientation discriminations, even in the presence of noise, but not when performance is limited by contrast.