Visual Sampling of In-Vehicle Text Messages: Effects of Number of Lines, Page Presentation, and Message Control

Emerging in-vehicle technology associated with radio data systems, satellite radio, navigation aids, and infotainment systems will expose drivers to an increasing amount of textual information. The visual demands of driving make it critically important to understand how characteristics of such information affect visual sampling of the roadway. This study used a medium-fidelity driving simulator to evaluate the effect of scroll control (manual or automatic), the number of lines displayed (one, two, or four), and scrolling strategy (line by line or page by page) on drivers’ visual sampling behavior of text messages. Fifteen men and 15 women aged 25 to 51 (mean 35.7, s.d. 7.5) participated. Consistent with previous models of sampling behavior, text message characteristics affected the number of glances much more strongly than the duration of glances, and drivers were able to protect driving performance even with a demanding secondary visual task. However, the number of lines of text displayed increased the mean glance duration, the variability of glance duration, and the number of glances greater than 2 s. Scrolling text line by line increased visual demand relative to scrolling page by page, particularly when scrolling is manually controlled by the driver. Manual control of scrolling by using a touch screen button led to greater visual demand compared with automatically scrolled text.