Divided multimodal attention: Sensory trace and context coding strategies in spatially congruent auditory and visual presentation

Abstract

Previous unimodal and multimodal research suggests that detection is a capacity-free process while identification is a capacity-limited process. The trace/context model assumes that different memory strategies are responsible for this difference, and not inherent differences between identification and detection. The model assumes that people use two strategies, a sensory trace or a context coding strategy and that if one is blocked, people will automatically use the other. Most previous experiments have presented lights and sounds in separate locations, creating the possibility of a spatial confound, which introduces an alternative interpretation of the results. This paper outlines a series of experiments, investigating divided multimodal attention, without the spatial confound, and challenges the assumption of the trace/context model, that only two strategies are available to participants. Our critical experiment involved a gap, which according to the trace/context model blocks the sensory trace strategy, simultaneously with a roaming pedestal which should block the context coding strategy. The results clearly show that people can use strategies, other than sensory trace and context coding; necessitating changes to the trace/context model.