Perceived exertion and performance modulation: effects of caffeine ingestion and subject expectation

It is well established that caffeine has ergogenic effects on endurance performance. This evidence often comes from studies in which subjects receive either caffeine or placebo in double-blind, randomized and counterbalanced order. Here, we propose a new methodology which aims to estimate the effects of participant expectancy of ergogenic or anti-ergogenic effects on performance.

Sixteen physically active participants (non-athletes engaged in systematic physical training >3 months, at least three times a week) performed three 30-minute running tests after being told they would be provided with either a harmful treatment (lactic acid), a beneficial treatment (caffeine) or a placebo. In each blinded case, subjects were given caffeine. The velocity and Rating of Perceived Exertion (RPE) during the time trial were examined in light of the participant's expectancy before and after the endurance event using Bayesian multilevel models.

For pre-exercise expectancy, there is a 92% probability that caffeine expectation decreases RPE (posterior median±SD -0.65±0.36) and a 79% probability that lactic acid expectation increases RPE (posterior median±SD 0.58±0.47) with expectations for placebo and 'not sure' falling in between (posterior median±SD: -0.37±0.32 and -0.22±0.37, respectively). In general, our interventions suggest an 81% probability that caffeine lowers RPE. However, there was no effect of caffeine supplementation on running velocity (median±SD 0.04±0.08 km.h^-1).

When a participant believed they are under a potentially positive treatment, their RPE decreased but if they believed themselves to be under a harmful treatment, their RPE increased, regardless of the actual positive intervention; neither caffeine nor the expectancy of a particular intervention improved actual performance as measured by running velocity in a 30-minute period.