Abstract
Attending to a visual stimulus feature selectively modulates activity in visual neurons, resulting in a space-independent perceptual enhancement of the attended feature (feature-based attention, FBA). We investigated whether holding the representation of a visual feature in working memory (WM) produces a similar effect. We further asked whether the effects of WM and FBA occur independently of each other. In three different experiments, seven subjects fixated a dot in the middle of a computer monitor while they identified the direction of a brief pulse of coherent motion in a random dot pattern (RDP) with 0% coherence located eccentrically. Simultaneously, they performed another task that required them to either: a) attend to the motion direction of a second RDP co-occurring with but far from the pulse (FBA-only Experiment); b) remember the motion direction of a previously presented RDP (WM-only Experiment); or c) both (a) and (b) concurrently (WM-FBA Experiment). We found that mean performance in the pulse direction identification task was significantly higher when either the attended direction (FBA-only Experiment, p=0.05, paired t-test), or the remembered direction (WM-only Experiment, p=0.004) was the same as the pulse direction, relative to when it was opposite. This suggests that visual features held in WM (likewise attended features) are preferentially perceived. In the WM-FBA Experiment, mean identification performance was highest when both the remembered and the attended directions were the same as the pulse direction, lowest when both were opposite, and intermediate when one was the same and the other opposite. A two-factor ANOVA revealed significant main effects of WM (p=0.04) and FBA (p=0.03), and a non-significant interaction between these two factors (p=0.25). Our results suggest that WM and FBA can independently modulate motion perception in humans.