Glance analysis of driver eye movements to evaluate distraction

Abstract

With the increasing use of in-vehicle devices in cars, an understanding of the safety implications of secondary tasks has become crucial. It is now possible to study the effects of many in-vehicle devices and tasks on driving by using head-mounted eye-tracking devices (HEDs) to collect eye positions and pupil diameters, which have been considered indicators of attentional focus. The collection of eyeposition and pupil-diameter data of automobile drivers under on-road conditions and while completing various secondary tasks is described in this paper. Drivers were asked to drive on a preselected twolane road for a total distance of 22 miles while gaze data were recorded using a HED. Longer off-road fixation durations were observed in radio-tuning and rearview mirror checking tasks, but not in the odometer checking task. In addition, the standard deviations of fixation displacements during a cognitive task involving the computation of a date for a meeting were shorter than those observed during normal driving.

Using self-reported data to assess the validity of driving simulation data

“…Although flight simulators are widely employed for pilot training, where important decisions on crew certification are often made, there is "no substantial body of knowledge to predict or measure the effectiveness of flight training devices" (Rolfe & Hampson, 2003). Similarly, human factors research has long used on-road and simulated driving tasks to assess driver behavior and performance in the presence of a variety of technological innovations (Ben-Yaacov, Maltz, & Shinar, 2002;Brown, Tickner, & Simmonds, 1969;Decina, Gish, Staplin, & Kirchner, 1996;Landau, Laur, Hein, Srinivasan, & Jovanis, 1994;McKnight & McKnight, 1993;Nunes & Recarte, 2002;Sodhi, Reimer, & Llamazares, 2002). There are two problems with the validity of measures from many driving simulation studies.…”

Miles away: determining the extent of secondary task interference on simulated driving

“…In combination with the results of previous studies (Kubose et al, in press;Patten et al, 2004;Sodhi et al, 2002) With a proliferation of driver information and entertainment systems being introduced into cars, it seems that vehicle designers must be ever more careful not to distract the driver on any attentional channel. However, there is some hope for accommodating technology while maintaining performance, in the shape of multisensory displays (Lansdown, 2001, Sarter, 2000 or integrated interfaces (e.g., Michon, 1993 .…”

“…Baber, 1991, Liu, 1996. Moreover, it is possible for a dual task experiment to compete for the same resource pool without adversely affecting the primary task (Baber, 1991); conversely, some studies have found that mental secondary tasks can affect vehicle control (e.g., Patten et al, 2004, Recarte and Nunes, 2002, Sodhi et al, 2002. Furthermore, there is evidence that there are other qualitative aspects of secondary tasks which can determine the extent of their interference effects, such as whether they are forced-pace or interruptible Kersloot, 2004, Noy, Lemoine, Klachon andBurns, 2004) .…”

The use of volumetric projections in Digital Human Modelling software for the identification of Large Goods Vehicle blind spots

“…There has been no research into the strategies that are used by LGV drivers to perform this task. Sodhi (2002) established that the mean 'eyes off road time' required to look at a Class III mirror (rear view mirror) in a passenger car is 0.96 seconds with a mean eye movement time of 0.32 seconds. To provide an indication of the difficulty that LGV drivers face these figures could be applied to the visual scanning of six mirrors resulting in a task time of 5.76 seconds.…”