1.   These results were replicated in two further laboratory studies using different judgment tasks and stimuli.

2.   The two judgment tasks were described to them before the drive as follows.

3.   First the results from the judgment tasks are reported and relationships between these judgments and a variety of objective measures are explored.

4.   Subjects generally found the judgment tasks relatively easy to perform and were able to give their first rating immediately upon hearing the tone in almost all cases.

5.   This indicates that there are no large overall differences in the judgment tasks dependent on the sex or annual mileage of the driver.

6.   The interaction between risk rating and accident estimate across the ten junctions suggests that subjects were able to sensibly dissociate the two judgment tasks in the laboratory setting.

7.   The only significant difference between the studies in judgment task was the main effect for accident estimate.

8.   The normality judgment task has an additional purpose.

9.   The normality judgment task was included to provide such a measure.