The costs of monitoring simultaneously two sensory modalities decrease when dividing attention in space.

Traditional views of multisensory integration emphasise the advantage of stimulating or attending to different senses at one single spatial location. We challenge this view demonstrating that in-parallel processing of two sensory modalities can be more efficient when attention is spatially divided rather than focused. We asked subjects to monitor simultaneously vision and audition either at one location (focused attention) or in the two opposite hemifields (divided attention) or to monitor one single modality at one or two locations. Behavioural results demonstrated that the costs of monitoring two modalities, versus one modality, decrease when spatial attention is divided between two separate locations compared with focused attention. Neuroimaging data revealed increased activity in the posterior-parietal cortex (PPC) when monitoring two modalities at different locations, while no specific region was recruited in the focused attention conditions. We suggest that supramodal control and the integration of spatial representations hinders the selection of independent sensory streams when attention is spatially focused, while a greater exploitation of modality-specific resources and the engagement of PPC allows in-parallel processing when attention is spatially divided. Copyright (c) 2009 Elsevier Inc. All rights reserved.