Conditioned effects of caffeine on performance in humans

There is limited evidence for the conditioning of stimulant-like drug effects to previously-neutral stimuli in humans. Two studies tested whether the facilitatory effects of caffeine on cognitive performance can be conditioned to the context of drug administration. In Experiment 1, sixteen participants were divided equally into two groups: one group (the "paired" group) received 250 mg caffeine in a novel beverage prior to completing two computerized performance tests; the other group (the "unpaired" group) received the same beverage without caffeine (i.e. placebo) before testing. After the performance tests, the unpaired group received the caffeinated drink, and the paired group received placebo in a different context from that in which testing had taken place. The performance tests comprised a test of simple reaction time to a visual stimulus and a test of logical reasoning (the semantic verification task). The procedure was repeated over four separate conditioning trials. On a subsequent fifth session, the test for a conditioned response, all participants received placebo before test. Simple reaction time was significantly reduced by caffeine over the four conditioning trials, and on the test for conditioned responding the paired group performed significantly faster than the unpaired group, indicating the development of a conditioned response. In Experiment 2, twelve participants attended four conditioning trials in which either 250 mg of caffeine (two sessions) or placebo (two sessions) were paired with two visually distinct environments. After completing the conditioning sessions, tests for conditional responding were conducted by administering placebo in both contexts. During the conditioning phase, caffeine significantly improved reaction time performance relative to placebo, and this advantage was maintained at test in the CS+ context when placebo was administered in both contexts. Therefore the facilitatory effects of caffeine on performance can be elicited, in the absence of drug, by previously-neutral contextual stimuli that have been paired with drug administration.