Placebo effects on cycling performance in virtual-reality and laboratory environments

Placebo effects are a desirable outcome resulting from a person’s expected and/or learned response to a treatment or situation In sport, most research has examined placebo effects by administering a placebo and informing athletes that they received an ergogenic aid, or via manipulating their expectations about an opponent. While previous research has revealed the magnitude of placebo effects during sport performance, it is limited in that they are often conducted in highly controlled environments, and opponents are often a replication of participants’ own performance. Thus, it is unknown if placebo effects are induced outside of the laboratory and whether they can be induced when competing against real opponents. In this research programme, placebo effects induced via both ergogenic aids and opponents were examined when participants completed cycling time trials remotely on a virtual-reality software (i.e., Zwift) or in the laboratory. In Study 1 (N = 44), the reproducibility of 20-min cycling performance on Zwift was confirmed (CV = 3.7%). In Study 2 (N = 67), athletes completed two 20-min cycling time trials on Zwift, before completing a final time trial with the administration of one of four conditions as part of the balanced placebo design [1) told beetroot/given beetroot, 2) told beetroot/given placebo, 3) told placebo/given beetroot, and 4) told placebo/given placebo]). Findings showed no differences in power output (ηp2 = .03) during any condition in comparison to baseline. In Studies 3 and 4, a deceptive intervention was adopted to investigate the effects of different competitive environments on cycling performance, whereby participants were either correctly informed about the nature of the opponent (accurate condition) or informed they received a performance-enhancing substance (deception). In Study 3, after a 20-min baseline time trial, participants (N = 12) competed twice against a virtual avatar replicating their previous baseline performance (competitionBSL) or against a virtual avatar riding at 2% higher power outputs than their best competitive performance (augmented feedback conditions; accurate and deception). Results showed that participants improved performance during competitionBSL (P < .001) and accurate (P = .036) in comparison to baseline but not during deception (P = .152). In Study 4 (n = 14), participants competed against a real opponent of similar ability (± 2% difference achieved during baseline). Contrary to Study 3, performance during both accurate and deception conditions was similar to baseline (all P ≥ 0.134). Collectively, this research has shown that placebo effects might not be as evident in remote-research designs than when conducted in the laboratory, which could be explained by the limited social contact between researchers and participants. These results have important implications for researchers and practitioners interested in placebo effects outside of the controlled environments, highlighting the importance of considering the exercise context. Virtual-reality software an innovative tool in which to conduct experimental designs in applied settings, including a geographically diverse sample, perhaps increasing the generalisability of findings.