In this paper two experiments are reported which were designed to investigate the effects of smoking on the performance of a rapid information processing task. The task involves the detection of sequences of three consecutive digits of the same parity from a series of digits presented visually at the rate of 100/min. In the first experiment smoking improved both the speed and accuracy of performance above rested baseline levels, the greatest improvement occurring with the highest nicotine and tar delivery cigarette. In the second experiment smoking again improved the speed and accuracy of performance above baseline levels, while performance deteriorated over time after not smoking as well as after smoking a nicotine-free cigarette. These findings demonstrate that smoking produces absolute improvements in performance and are explained in terms of the action of nicotine on central cholinergic pathways.