The effects of display context on the effectiveness of visual onsets for attention capture

Breakdowns in human-automation communication have been observed in many complex domains for a variety of reasons. For example, numerous studies have shown that pilots on highly automated aircraft sometimes miss unexpected changes in autoflight modes. These changes are signaled by the onset of a green box drawn around the new mode name on the primary flight display. The purpose of the appearance of the green box is to capture pilots' attention in case of a mode transition. However, the above difficulties suggest that this design choice may not be effective in the context of data-rich multiple-dynamic display contexts. The goal of the current experiment was to examine systematically the effectiveness of visual onsets for capturing attention when these onsets are immersed in several different display backgrounds that vary with respect to color and dynamics. Detection performance for embedded targets was worst in the so-called color-dynamic condition, which resembled many real-world displays. The other display conditions suggest that dynamics affect performance more strongly than color similarity between target and background. The results of this study illustrate the need for prudent adaptation and application of findings from laboratory research on attention capture to the design of interfaces for complex systems and environments.