A Meta-Analysis of Driver Eye-Scanning Behavior While Navigating

Abstract

A meta-comparison of link analyses was performed using two eye scanning studies. One study was an analysis of the Etak navigator performed by Antin, Dingus, Hulse and Wierwille, (1990). Antin and his associates compared the Etak system to a paper map control condition and a memorized route baseline condition. The second study was an evaluation of the TravTek system performed by Dingus, McGehee, Hulse, Jahns, Natarajan, Mollenhauer and Fleischman (1995). The TravTek Camera Car Evaluation tested six different navigation conditions (a TravTek visual display of a full, heading-up route map with voice guidance, the same TravTek route map without voice guidance, a TravTek visual display showing a graphic representation of static turn-by-turn information, the same turn-by-turn screens without voice guidance, a textual paper direction list with large legible font, and a conventional paper map). The eye glance data shown in the comparison of these two studies revealed several interesting findings. In all conditions, scanning of instruments, mirrors and signs/ landmarks was a low frequency occurrence and largely constant. Increases in visual attention by a navigation condition draws attention from forward, left and right roadway scanning resources. Thus, navigating draws upon potentially valuable accident avoidance resources in some circumstances. The eye scanning results also shared the relative benefits of the addition of a highlighted route to a moving map display, the simplification of a visual display from a full map to a turn-by-turn graphic, and the addition of a voice supplement to a navigation aid.