Auditory-visual interaction and the timing of inputs

Summary

The objective of this study was to analyze the structural properties of the respective inputs that are conducive to immediate auditory-visual cross-modal bias. The study was designed as an updated and extended replication of one by Thomas (1941). Subjects were presented with either auditory or visual target signals in several positions around the median plane, together with a competing signal, always in the other modality, 15° to the left or to the right of that plane. The task was to indicate whether the target signal came left or right of center. Target and competing signals were delivered according to three temporal configurations: continuously on for 4 s, or periodically interrupted at either a fast or a slow tempo, and all combinations of the three configurations were used. Judgements of the location of the auditory target signal were attracted toward the visual competing signal in all conditions but two, those with a periodic target signal and a continuous competing one. Conditions with the two signals in the same configuration yielded larger biases than those combining different configurations, confirming that synchronization of discordant inputs is a major condition of cross-modal interactions. The occurrence of significant bias in nonsynchronous conditions, on the other hand, suggests that another factor might be the attraction of localization responses by competing signals with salient temporal configurations, and that interruption might be one important source of saliency. Auditory biases of visual localization were, as usual, smaller than visual biases, but nevertheless reached significance in a majority of conditions, and were also influenced by timing in much the same way as the latter.