Item

Abstract

Multisensory integration is thought to be mediated by the activity of superior colliculus (SC) and to be different from unisensory integration (i.e. the integration of two stimuli of the same modality). To verify these hypotheses, six healthy subjects, six patients with subcortical lesions not involving SC and six patients with collicular lesions were tested in a redundant signals paradigm. Subjects were presented with modality-specific stimuli (visual or auditory), cross-modal (audio-visual) and within-modal (visual-visual) stimulus pairs. In all the three groups, a significant reduction in reaction times (redundant target effect: RTE) was found with audio-visual stimuli, compared to modality-specific stimuli. No significant difference was evident with within-modal pairs. The RTE for audio-visual stimuli observed in healthy subjects and patients with subcortical lesions not involving SC, violating the race model inequality, was explained by a neural coactivation mechanism. In contrast, in patients with collicular lesions, no violation of the race model was found and the RTE was explained by a statistical facilitation mechanism. Overall, these results suggested that multisensory integration is more effective in speeding responses compared to unisensory integration and is strictly dependent on the activity of SC, since patients with collicular lesions failed to demonstrate multisensory integrative abilities.