Cannabis Users Felt Ready to Drive Before Simulator Performance Recovered
Feeling increasingly ready to drive after cannabis did not track recovery in a controlled simulator study. At 90 minutes after smoking, willingness to drive had increased substantially while the THC group’s driving performance remained worse than placebo.
What participants actually did
The study analyzed 191 regular cannabis users assigned to placebo or cigarettes containing 5.9% or 13.4% THC. Participants smoked to their usual desired level of intoxication rather than receiving a fixed absorbed dose. Driving simulations occurred before smoking and approximately 30 minutes, 90 minutes, 3.5 hours and 4.5 hours afterward.
Before each drive, researchers asked participants about perceived impairment and whether they would drive in their current state. The composite simulator score separately incorporated lane control, speed variability, divided attention and car-following performance.
Confidence improved before measured performance
Among THC participants answering the question, 47.5% would drive at 30 minutes; 68.6% would do so at 90 minutes. Yet the THC group still showed a significant performance decrement relative to placebo at both times.
This mismatch matters because deciding to drive is itself a behavioral outcome. A person may recognize some drug effect while becoming more willing to accept it. Neither a declining feeling of being high nor willingness to drive independently verifies recovery.
Later null results do not create a personal clearance time
The group difference was not statistically significant at 3.5 hours or 4.5 hours. No simulator measurements occurred between 90 minutes and 3.5 hours, so the study cannot identify an exact recovery point within that interval.
The selected regular-user sample, self-directed smoking and simulator setting limit generalization to infrequent users, edibles, repeated consumption or longer drives. A nonsignificant group comparison also does not establish identical performance in every participant.
Keep self-report, toxicology and function separate
Whole-blood THC concentrations did not correlate significantly with the composite driving score at the assessed times. Thus, confidence and concentration each supplied information different from measured performance.
The practical lesson is to document when confidence was reported and when functional observations were made. Related publications from this trial analyze the same research program; they should not be counted as independent replications of this confidence–performance mismatch.
Related: Interpreting THC concentration and impairment.
Primary research
Driving Performance and Cannabis Users’ Perception of Safety: A Randomized Clinical Trial. DOI: 10.1001/jamapsychiatry.2021.4037.
