Measuring visual demand of mobile devices while driving - An application of the occlusion method

Recent years have seen a strong expansion of computing technology beyond the traditional office-based desktop tasks. By this portable devices have become a permanent companion in our daily life and culture including the driving context. More and more portable devices, such as handhelds, PDAs, can be equipped with software that provides the driver with traffic and navigation information. Being able to use these portable devices in the car provides the driver with clear benefits: Relevant information can be entered into the system before the driver enters the car and is available while driving. A PDA provides the driver with all information about oncoming meetings and the day's time schedule previously entered into the PDA's calendar function while being at the office. Nevertheless, there is some concern that the use of such devices while driving can distract the driver in a way that leads to a significant reduction in the driver's safety (Commission of the European Communities. (2006). Commission recommendation of 22 december 2006 on safe and efficient in-vehicle information and communication systems: Update of the european statement of principles in human machine interface: Available at http://www.esafetysupport. org/download/working-groups/hmi-esop221206-cn.pdf). One of these concerns is related to driver distraction caused by the visual demand these devices impose on the driver. To avoid the design of HMls that arc too visually demanding it is necessary that measurement procedures arc available that allow to assess the visual demand of HMls. One of these procedures is the so called "Occlusion Method". This method consists of repeatedly presenting and occluding the HMI of a system while a person performs a task with this HMI. The basic idea behind this method is: The more visually demanding the task with the given HMI is the less able is the person to perform the task while the HMI is occluded and the longer it takes to complete the task, therefore. The less visually demanding the task is the more able is the person to use the occluded time intervals to perform the task and the smaller is the increase of task completion time compared to task completion time without occlusion. The aim of this study was to use this method to assess the visual demand of typical HMls of different mobile devices that are used while driving. 22 participants between 40 and 58 years of age performed 6 different tasks. Each task was performed by each participant alone (baseline), with the sight on the HMI repeatedly occluded (occlusion), and while driving in the driving simulator. For each task the occlusion method assessment was calculated and compared with different measures of driving performance. The results show a high correlation between the occlusion index and certain parameters of driving performance such as the standard deviation of the lateral position or the number of lane departures while task performance. These results indicate that the occlusion method seems to be a valid procedure to assess the visual demand of tasks performed with HMls of mobile devices and to predict the effects of such tasks on driving performance.