OLDER DRIVER SIMULATOR BASED INTERSECTION TRAINING: THE EVALUATION OF TRAINING EFFECTIVENESS AND SIMULATOR SICKNESS
Michael A. Knodler
Older drivers are over involved in intersection crashes. The evidence to date suggests that this is primarily because they fail to look for potential threat vehicles after they enter a stop-controlled intersection. These secondary glances are absolutely critical when the built or natural environment obscures such vehicles while the driver is stopped before entering the intersection. Simulator-based older driver training programs exist which double the frequency of secondary glances that older drivers take up to two years after training. However, almost 40% of those who participate in such training never finish because of Simulator Adaptation Syndrome (SAS, or “simulator sickness”). Two factors are believed to contribute to the high simulator sickness rates: 1) the relatively high immersion at each point in time and 2) the relatively long period of time over which the training occurs in the simulator. In this experiment, simulator micro-scenarios were designed to train older drivers to take secondary looks. These micro-scenarios take no more than 30 to 45 seconds to complete and are much shorter than the 20 minute training programs now available. In addition, level of immersion was varied, from relatively low (the virtual world was projected onto three 22'' diagonal LCD monitors) to medium (the virtual world was projected onto one to three 60'' screens). A total of five groups of older drivers (91 total between the ages of 67 and 86) were run in the experiment. Three of the groups were given active, secondary glance training on a driving simulator -- one on a low-immersion simulator and two on a medium-immersion simulator (one group utilized all three screens and one group utilized only one screen) -- one of the groups was given passive, secondary glance training using a PowerPoint presentation and one of the groups received no training at all, control group. After the training was delivered participants in all five groups were evaluated in the field while driving alone in their own vehicle as they wore a head mounted camera. Secondary glances were recorded from the videos of the drives captured by the camera. The simulator training dropout rate was reduced radically from what has been reported in the literature (roughly 40%), to 14.3% in the three screen medium-immersion simulator, 6.3% in the one screen medium-immersion simulator and 11.8% in the low-immersion simulator. The percentage of secondary glances in the field increased significantly for the group given active, 3-screen medium-immersion simulator training (82%) above those given passive, PowerPoint training (69%) and those who received no training, control group (42%). There was no statistically significant difference between the group given active, low-immersion simulator training (74%) and the group given passive, PowerPoint training; however, statically significance exists between the three active training groups and the 1-screen medium immersion simulator training (58%). It is clear that the design of micro-scenarios in a lower immersion environment decreased simulator sickness and increased the frequency of secondary glances.