Abstract
Vigilance is the ability to sustain goal-directed attention over long periods of time. Previous research shows the ability to maintain performance may decline during solely visual and solely auditory vigilance tasks, resulting in misidentified stimuli, i.e. the vigilance decrement. However, little research has compared the vigilance decrements in single sensory modality tasks to those that may occur under multimodal conditions, such as audiovisual tasks. To address this knowledge gap, we tracked stimulus identification rate using a modified version of the Integrated Visual and Auditory Continuous Performance Test across three vigilance conditions: visual, auditory, and audiovisual. This task required participants to monitor for a stimulus in the presence of distractors. The task stimuli were modified to be “b” and “p” instead of “1” and “2”, because these letters are difficult to discriminate between in visual and auditory modalities. To prevent ceiling effects on performance, the auditory stimuli were masked with static noise whereas the visual stimuli had shortened tails to appear more similar. The stimulus-to-distractor ratio was set to 1:5.25. Each session included four blocks of 250 trials which took approximately 26 minutes. One hundred subjects were recruited using Amazon MTurk. Differences in stimulus identification rate and reaction time were determined using 3X4 mixed ANOVAs with the between-subjects factor of condition and the within-subjects factor of block. Results indicated a significant interaction between condition and block such that stimulus detection rate significantly decreased in the audiovisual group relative to the other conditions. There were no systematic changes in reaction time, suggesting that the decrease in accuracy was not due to a speed-accuracy trade-off. These results suggest that monitoring multiple modalities of information versus a single modality results in a larger vigilance decrement, and thus may be more cognitively taxing.