The Automatic conditions for both Voice and Screen displays produced longer forward glances (and larger overall allocation of visual attention) than the Interactive conditions.
Figures 17.1-17.5 illustrate the main results of the two experiments. Means and standard error bars are shown for quality of driving, time load index, percent attention allocation to the forward road scene, percent attention allocation to the in-vehicle display, and navigation errors as functions of navigation task conditions. The conditions S/A, S/I, V/A, V/I, PL, IN, and Control refer to Screen-Automatic, Screen-Interactive, Voice-Automatic, Voice-Interactive, Paper List, Ideal Navigator, and Familiar Route navigation conditions, respectively.
The figures give a consistent picture of the influence of different attributes of a navigation system on task demands and driver behavior. Following a familiar route without searching for addresses (the control condition) was the least demanding task for drivers, as expected. Their visual attention