Attentional load impacts multisensory integration, without leading to spatial processing asymmetries

The present study examined whether spatial processing in the unimpaired cognitive system is influenced by attentional load during multitasking. More specifically, it tested the hypothesis that high attentional load would induce spatial processing asymmetries in the form of a rightward attentional bias. We conducted two separate experiments on healthy adults (n = 101 and n = 98) by using web-based data collections. We capitalized on a condition of perceptual uncertainty to investigate the presence of these spatial asymmetries which cannot be easily detected under regular perceptual conditions. More specifically, we employed a primary audiovisual integration task, which involved presenting stimuli capable of eliciting the sound-induced flash illusion (i.e., task-relevant flashes accompanied by an incongruent number of sounds) on either the left or right side of the screen. This task enabled us to investigate audiovisual integration, but also indirectly provided an opportunity to sensitively explore spatial processing within a highly complex context. In Experiment 1, attentional load was increased by presenting stimuli to be retained before the audiovisual integration task (i.e., "offline" attentional load manipulation). Differently, in Experiment 2, attentional load was increased by having participants to perform visual discrimination during the audiovisual integration task (i.e., "online" attentional load manipulation). Attentional load was increased in a different way within each experiment to test the idea that more demanding tasks, albeit of different nature, would have similarly modulated performance. In both experiments, we replicated the increase of sound-induced flash illusion under high attentional load, which challenges the notion of an early and pre-attentive onset of the illusion. However, this effect was identical for left- and right-sided flashes, which speaks against the existence of load-induced spatial processing asymmetries in the unimpaired cognitive system. Given that both experiments yielded similar results, quantitative aspects of attentional engagement rather than the nature of the attentional resources involved seem to play a critical role.