The road crash potential of cannabis use has been investigated by means of several surveys of the reported accidents and violations of users compared with non-users. Results have been conflicting and it has been suggested that such differences as have been found may be explained by differences between users and non-users excluding those caused by temporary states of intoxication.

In the present study the driving performance of subjects who reported regular use of cannabis was compared with that of a group of non-using controls matched for age, education and driving history. Subjects were tested in a closed-course driving situation involving perceptual/decision making tasks and high speed driving. In addition to these measures the subjects' driving on the road was monitored surreptitiously, by activation of a video tape recorder situated within the car, while the subject drove the car to the experimental track.

The measures of driving performance in the closed-course and normal traffic situation showed some evidence for consistent driving patterns operating in both environments, particularly in the use of the vehicle controls and the speed of driving.

The cannabis users did not differ from nonusers in their use of vehicle controls, speed of driving or performance measures obtained in the closedcourse situation. There was a difference however, in overtaking behaviour and indicator use in the normal traffic situation suggesting some differences may exist in the non intoxicated driving pattern of cannabis users compared with non-users.