Abstract
Visual masking is a powerful methodological tool to control object visibility at different stages of sensory processing. An important question to ask is how attention modulates identified distinct processes and thus alters the perceived visibility of a target. Particularly, paracontrast masking paradigms provide an important framework to shed light on this basic question. Previous studies indicated that distinct processes are involved in paracontrast masking (Breitmeyer et al., 2006). For example, brief and prolonged inhibitions have been observed at short and long stimulus onset asynchronies (SOAs), respectively. They have been interpreted as reflecting low- and high-level inhibitory (i.e., lateral vs. recurrent) mechanisms in the parvo-dominated pathway associated with perceived visibility. In the current study, we specifically aimed to understand how spatial attention alters these inhibitory mechanisms. Similar to previous research on metacontrast masking (Agaoglu et al., 2016), we manipulated spatial attention by changing set-size in the visual field. We employed a contour discrimination task on the visual target under different set-size and SOA conditions. A two-way repeated-measures ANOVA on the normalized performance values revealed a significant main effect of SOA and a two-way interaction between SOA and set-size. To elucidate the source of two-way interaction, we performed additional post-hoc comparisons across set-sizes. A significant effect of set-size was present at both short and long SOAs over which the brief and prolonged inhibitions were observed, respectively. Interestingly, these attentional modulations were in the opposite direction such that an increase in set size decreased the brief inhibition while increasing the prolonged inhibition. To conclude, these results indicate that spatial attention can take place at both low- and high-level visual processing. More specifically, they point to the differential effects of attention on the inhibitory mechanisms involved in controlling stimulus visibility at different processing stages.