ABSTRACT
Interface designers are increasingly using auditory, tactile, and/or multimodal warning signals to capture an operator’s attention. We report an experiment showing that multimodal warning signals can sometimes be no more effective in than unimodal signals in capturing a person’s attention, and may actually be somewhat worse under certain conditions. A potentially fruitful approach to predicting the effectiveness of multimodal warning signals relies on an understanding of the neurophysiology of those brain structures that integrate multiple sensory cues for the control of orienting. Behavioural and neurophysiological data both suggest that better multimodal warning might be achieved by appropriate temporal desynchronization of the component stimuli, to mimic the environmental cues we are optimally suited to detect.
