Temporal uncertainty determines the use of object representations in attentional guidance

EXPRESS: Contextual uncertainty of visual scene modulates object-based attention

Contextual information involves invariant properties that are critical in selective attention. There is no direct evidence showing the effect of contextual information on object-based selective attention. The current study aimed to investigate the role of contextual uncertainty on object-based effect using a flanker task and to clarify the contradictory results obtained in previous studies. Herein, contextual uncertainty specifically referred to the configurations of the stimuli presented randomly as vertical or horizontal displays (high contextual uncertainty) that was reduced by showing consistent configurations within a block, via implicit learning of configuration (low contextual uncertainty). In Experiment 1, the object-based effect was observed under the high uncertainty condition and disappeared under the low uncertainty condition, demonstrating that contextual uncertainty modulated object-based attention. Experiment 2 provided explicit knowledge of the configural orientations, which can be utilised to sufficiently guide subsequent perception with increase in cueing interval, and therefore, affected contextual uncertainty. Relative to a short cueing interval, the long cueing interval enabled the participants to utilise the contextual knowledge for guiding visual attention and reducing uncertainty. Consistent with the finding in Experiment 1, the explicit manipulation of contextual uncertainty affected the object-based effect. The results proved that contextual uncertainty played an important role in prioritisation in object-based attentional selection. The mechanism of the interplay between contextual uncertainty and object-based attention was discussed.

Prioritization to visual objects: Roles of sensory uncertainty

Two hypotheses, attentional prioritization and attentional spreading, have been proposed to account for object-based attention. The attentional-prioritization hypothesis posits that the positional uncertainty of targets is sufficient to resolve the controversy raised by the competing attentional-spreading hypothesis. Here we challenge the sufficiency of this explanation by showing that object-based attention is a function of sensory uncertainty in a task with consistent high positional uncertainty of the targets. In Experiment 1, object-based attention was modulated by sensory uncertainty induced by the noise from backward masking, showing an object-based effect under high as compared to low sensory uncertainty. This finding was replicated in Experiment 2 with increased task difficulty, to exclude that as a confounding factor, and in Experiment 3 with a psychophysical method, to obtain converging evidence using perceptual threshold measurement. Additionally, such a finding was not observed when sensory uncertainty was eliminated by replacing the backward-masking stimuli with perceptually dissimilar ones in Experiment 4. These results reveal that object-based attention is influenced by sensory uncertainty, even under high positional uncertainty of the targets. Our findings contradict the proposition of attentional spreading, proposing instead an automatic form of object-based attention due to enhancement of the perceptual representation. More importantly, the attentional-prioritization hypothesis based solely on positional uncertainty cannot sufficiently account for object-based attention, but needs to be developed by expanding the concept of uncertainty to include at least sensory uncertainty.