Abstract
Recent developments of in-vehicle information systems provide a lot of
information to the drivers. Despite the usefulness of these systems, it is of great concern
that visual attention demanded by them leads to driver distractions. Therefore, this study
proposes a new method to evaluate the influence of distractions to drivers caused by
information processing demand of in-vehicle information systems. Our method focuses on
two types of involuntary eye movement of the driver: one is VOR (vestibulo-ocular reflex)
and the other is OKR (optokinetic reflex). These two types of eye movement occur reflexly
for gaze stabilization while driving. We investigated the influence of distractions on VOR and
OKR using our proposed model-based method. The first step of the method is to identify the
eye movement model for a particular driver from experimental data measured without any
secondary task. This model represents the dynamics of VOR and OKR. After that the eye
movement is measured when the driver get distracted by paying attention to secondary tasks,
and it is compared with the identified model output. This method makes it possible to detect
the influence of distractions on VOR and OKR. This study investigated the influence of
distractions on driver eye movement by varying the amount of information processing
demand of secondary task. Consequently, we confirmed the possibility of quantitatively
evaluating driver distractions using our proposed method.
Keywords - Driver distractions, Information processing demand, Reflexive eye
movement, Eye movement modeling, State estimation