Summary Aim of the study To explore the effects of caffeine and bright light therapy on simulated nighttime driving in sleep-deprived healthy volunteers. Participants and methods Twelve male healthy volunteers aged 20 to 50 years participated in a randomized cross-over study of simulated nighttime driving at a sleep laboratory, followed by recovery sleep with polysomnography at home. The volunteers received variable combinations of caffeine 200mg (C+), caffeine placebo (C–), bright light 10,000 lux (L+), and bright light placebo<50 lux (L–), in four sessions (C+L+, C+L–, C–L+, C–L–), in random order with a wash-out period of 7 days. Treatments were given at 1 a.m. and testing was performed at 1:30 a.m., 3 a.m., 4 a.m., and 6 a.m. Lane drifting was the primary outcome measure. Other measures were reaction times, self-rated fatigue, sleepiness and recovery sleep. Results Without treatment, lane drifting increased throughout the night, and objective and subjective vigilance declined. Paired comparisons showed that lane drifting was significantly worse at 6 a.m. and at 4 a.m. than at 1:30 a.m. There was a global treatment effect on lane drifting. Lane drifting at 6 a.m. was significantly decreased with C+L+ compared to C–L–. Conclusions Bright light therapy combined with caffeine administered at 1 a.m. decreased lane drifting by healthy volunteers during simulated nighttime driving.