The Effects of Task Irrelevant Images on Glance Time for In-vehicle Systems

Abstract

The goal of the present study was to understand how buttons and images affect glance times for in-vehicle tasks. Search tasks that simulated in-vehicle tasks (e.g., radio tuning, navigation) were used to measure these effects. Participants across various age groups were required to select an item (e.g., "Mango") on each screen that matched a category of items (e.g., "Fruit) that was given at the beginning of the task. For example, a category of "fruit" would have targets of "strawberry" or "mango." Glance times were collected for each screen to analyze the effects of the number of buttons and presence of images on the display. Additionally, the image detail was varied by using a generic graphic (e.g., polygons) and a photographic image. A regression model showed that the image variable resulted in a small improvement to the "number of buttons" model (∆R 2 = .005) but adding the "level of detail" variable (i.e., polygons or photograph) did not. However, the significant effect of images reflected a decrease from the condition with no images. Further analysis showed that the screens with the images had similar variability to the ones without. These results suggest that the effects of task irrelevant images are measurable and should be thought of as similar to other interface elements. Additionally, there does not appear to be a quality in images that causes higher glance time over and above other interface elements. The implications for distraction policy are discussed.