Multisensorische Integration von redundanten Reizen

If participants simultaneously receive two target stimuli of different modalities (e.g. auditory and visual), they respond to them faster than would be expected from their reaction times to simple stimuli (redundant target effect, RTE). This speeding of reaction time indicates that the information of the two sensory channels is integrated at a particular processing stage. The present thesis investigates this reaction time gain in five experiments in which the number of redundant targets, sequence effects, the spatial relationship and spatial attention were varied systematically. In addition to reaction time analyses, event-related potentials were measured in two experiments. In chapter 1, a new method to investigate redundancy gains in the reaction times to trimodal auditory-visual-tactile stimuli is described. Especially it is shown that redundancy gains in trimodal stimuli can entirely be explained within the framework of bisensory interactions. Responses to ipsimodal stimuli (e.g. an auditory stimulus following another auditory stimulus) are faster than to crossmodal stimuli (e.g. an auditory stimulus following a visual stimulus). Since the modality of at least one component of a bimodal stimulus always matches the modality of the preceding stimulus, it could be argued that bimodal stimuli are always ipsimodal. In chapter 2, it is demonstrated that this can yield artificial redundancy effects, and a method to avoid this potential problem is described. A frequent approach to study interactions of the auditory and the visual system with event-related potentials (ERPs) is to measure the ERP to auditory-visual stimuli (AV) and to compare it with the sum of the ERPs to auditory and visual stimuli (A+V). A problem of this ERP comparison is that the three ERPs should be free of common activity. In chapter 3, I describe an alternative comparison which is robust with respect to common activity. Does the spatial relationship of the two components of a bimodal stimulus affect the way in which the information of the two sensory channels is integrated? In chapter 4, it is shown that processing of spatially congruent redundant stimuli is more efficient than if the two stimuli are presented at different locations. Moreover, ERPs to spatially congruent bimodal stimuli differed from the ERPs to spatially incongruent stimuli, at parietal recording sites. This indicates that polymodal areas in the parietal cortex might be involved in the processing of bimodal stimuli. In chapter 5, it is shown that the redundancy gain depends on whether the participant is focusing the location of the bimodal stimulation. This indicates that multisensory integration of redundant stimuli is not only a stimulus-driven 'bottom up' process. It seems rather to occur at higher levels of processing, as suggested by recent models of visual information processing.