Hemispheric asymmetry: Looking for a novel signature of the modulation of spatial attention in multisensory processing

Abstract

The extent to which attention modulates multisensory processing in a top-down fashion is still a subject of debate among researchers. Typically, cognitive psychologists interested in this question have manipulated the participants' attention in terms of single/dual tasking or focal/divided attention between sensory modalities. We suggest an alternative approach, one that builds on the extensive older literature highlighting hemispheric asymmetries in the distribution of spatial attention. Specifically, spatial attention in vision, audition, and touch is typically biased preferentially toward the right hemispace, especially under conditions of high perceptual load. We review the evidence demonstrating such an attentional bias toward the right in extinction patients and healthy adults, along with the evidence of such rightward-biased attention in multisensory experimental settings. We then evaluate those studies that have demonstrated either a more pronounced multisensory effect in right than in left hemispace, or else similar effects in the two hemispaces. The results suggest that the influence of rightward-biased attention is more likely to be observed when the crossmodal signals interact at later stages of information processing and under conditions of higher perceptual load-that is, conditions under which attention is perhaps a compulsory enhancer of information processing. We therefore suggest that the spatial asymmetry in attention may provide a useful signature of top-down attentional modulation in multisensory processing.

Keywords: Asymmetry; Crossmodal attention; Hemispace; Multisensory integration; Prior entry; Rightward-biased attention.

Fig. 1

Schematic representation of the multisensory prior-entry effect reported by Spence et al. (2001). The proportion of “touch first” responses increased as a function of stimulus onset asynchrony (SOA; negative values indicate the visual-leading conditions, and positive values, the tactile-leading conditions), giving rise to a cumulative curve. In this curve, the 50 % “touch first” responses point corresponds to the SOA at which participants presumably perceived the two stimuli as simultaneous (i.e., the point of subjective simultaneity [PSS]). When one of the stimuli happened to be attended, it would be perceived as having been presented earlier in time than the other stimulus, thus leading to a shift of the PSS. Here we demonstrate an example in which the visual stimulus was presented on the right side and the tactile stimulus on the left (one of the conditions in their Exp. 4). The gray shading represents how the participants’ spatial attention was distributed. As compared to the baseline condition in which the participants divided their attention between both sides (black solid line), the shift of the PSS was smaller when the participants attended to the right (gray dashed line; the visual stimulus was attended, and therefore the PSS shifted toward the tactile-leading condition) than when the participants attended to the left (gray dotted line; the tactile stimulus was attended, so the PSS shifted toward the visual-leading condition). V, visual stimulus; T, tactile stimulus

Fig. 2

Mean accuracies of participants’ letter identification performance in Chen and Spence’s (2011) Experiment 1. The central panel represents the results combining the two hemispaces that were reported by Chen and Spence. The right and left panels represent the results of a reanalysis of the data, separating those conditions in which the visual target was presented in right or left hemispace, respectively. The error bars represent ±1 standard error of the means

Fig. 3

Mean accuracies of participants’ letter identification performance in Chen and Spence’s (2011) Experiment 5. The central panel represents the results combining the two hemispaces that were reported by Chen and Spence. The right and left panels represent the results of a reanalysis of the data separating those conditions in which the visual target was presented in right or left hemispace, respectively. In this experiment, the spatial consistency between the visual target and the sound was manipulated. The error bars represent ±1 standard error of the means

Fig. 4

Schematic figure capturing how spatial attention modulates multisensory integration. Darker gray shading represents a greater probability of attentional distribution/modulation. In the pathways of human information processing, multisensory signals interacting at a later rather than an earlier stage would be more susceptible to the top-down modulation of attention. In addition, attention is likely to be biased toward the right hemispace when the task loading is high. Taken together, the hypothesis of asymmetrical attentional modulation is that multisensory processing that requires attention would be more pronounced when the stimuli are presented in right rather than left hemispace (the upper-right corner). V, visual sensory input; A, auditory sensory input; T, tactile sensory input