The purpose of this study was to investigate the possibility of a placebo effect from carbohydrate (CHO) intake during prolonged exercise. Ten endurance-trained male cyclists performed 3 experimental trials consisting of 120 min of steady-state cycling at 61% VO2max followed by a time trial (TT) lasting approximately 60 min. During exercise participants ingested either plain water (WAT), artificially colored and flavored water (PLA), or a 6% carbohydrate-electrolyte solution (CES). PLA and CES were produced with identical color and taste. To investigate the possibility of a placebo effect from CHO intake, participants were told that both flavored solutions contained CHO and that the purpose of the study was to compare CHO drinks with water. Mean power output during TT was 218 +/- 22 W in WAT, 219 +/- 17 W in PLA, and 242 +/- 27 W in CES. Performance times were 66.35 +/- 6.15, 65.94 +/- 5.56, and 59.69 +/- 2.87 min for WAT, PLA, and CES, respectively. Therefore, CES ingestion enhanced TT performance by 11.3% compared with WAT (p < .05) and 10.6% compared with PLA (p < .05), with no difference between PLA and WAT. In conclusion, during a prolonged test of cycling performance, in which participants were not fully informed of the test conditions, there was no placebo effect when participants believed they had ingested CHO. In contrast, the real effect of CHO intake was a 10.6% improvement in TT cycling performance.