Cross-modal cueing in audiovisual spatial attention

Abstract

Visual processing is most effective at the location of our attentional focus. It has long been known that various spatial cues can direct visuospatial attention and influence the detection of auditory targets. Cross-modal cueing, however, seems to depend on the type of visual cue: facilitation effects have been reported for endogenous visual cues while exogenous cues seem to be mostly ineffective. In three experiments, we investigated cueing effects on the processing of audiovisual signals. In Experiment 1, we used endogenous cues to investigate their effect on the detection of auditory, visual, and audiovisual targets presented with onset asynchrony. Consistent cueing effects were found in all target conditions. In Experiment 2, we used exogenous cues and found cueing effects only for visual target detection, but not auditory target detection. In Experiment 3, we used predictive exogenous cues to examine the possibility that cue-target contingencies were responsible for the difference between Experiment 1 and 2. In all experiments, we investigated whether a response time model can explain the data and tested whether the observed cueing effects were modality-dependent. The results observed with endogenous cues imply that the perception of multisensory signals is modulated by a single, supramodal system operating in a top-down manner (Experiment 1). In contrast, bottom-up control of attention, as observed in the exogenous cueing task of Experiment 2, mainly exerts its influence through modality-specific subsystems. Experiment 3 showed that this striking difference does not depend on contingencies between cue and target.