Abstract
Sixty-five elderly drivers participated in an experiment to develop a risk estimation test that can measure self-monitoring ability. The test consisted to point hazards, blind spots, signals and road sign by touching display in a Three-Dimensional Computer Graphics (3DCG) image from the driver's viewpoint on a 21-inch touch panel display. After the test, participants were asked to self-evaluate the rate of the total objects they could touch for safety, and to select a velocity for their next driving. The results were compared with those of non-elderly drivers. And the relationship with cognitive function measured by the TMT-B was also examined. In addition, the variables affecting velocity selection were analyzed. The results revealed the following points. The hazard pointing rate of elderly drivers was significantly lower than that of non-elderly drivers. There were many elderly drivers whose self-estimation of the hazard pointing rate was high, even though the actual rate was low. There was a significant correlation between the decline in cognitive function and the decline in the potential hazard like blind spots at intersections pointing rate. There was no relationship between cognitive function or self-estimation of hazard pointing rate and selection velocity.
