Affecting Effects on Affect: The Impact of Protocol Permutations on Affective Responses to Sprint Interval Exercise; A Systematic Review and Meta-Analysis of Pooled Individual Participant Data

Abstract

Responses to sprint interval exercise (SIE) are hypothesized to be perceived as unpleasant, but SIE protocols are diverse, and moderating effects of various SIE protocol parameters on affective responses are unknown. We performed a systematic search to identify studies (up to 01/05/2021) measuring affective valence using the Feeling Scale during acute SIE in healthy adults. Thirteen studies involving 18 unique trials and 316 unique participant (142 women and 174 men) affective responses to SIE were eligible for inclusion. We received individual participant data for all participants from all studies. All available end-of-sprint affect scores from each trial were combined in a linear mixed model with sprint duration, mode, intensity, recovery duration, familiarization and baseline affect included as covariates. Affective valence decreased significantly and proportionally with each additional sprint repetition, but this effect was modified by sprint duration: affect decreased more during 30 s (0.84 units/sprint; 95% CI: 0.74–0.93) and 15–20 s sprints (1.02 units/sprint; 95% CI: 0.93–1.10) compared with 5–6 s sprints (0.20 units/sprint; 95% CI: 0.18–0.22) (both p < 0.0001). Although the difference between 15–20 s and 30 s sprints was also significant (p = 0.02), the effect size was trivial (d = −0.12). We observed significant but trivial effects of mode, sprint intensity and pre-trial familiarization, whilst there was no significant effect of recovery duration. We conclude that affective valence declines during SIE, but the magnitude of the decrease for an overall SIE session strongly depends on the number and duration of sprints. This information can be applied by researchers to design SIE protocols that are less likely to be perceived as unpleasant in studies of real-world effectiveness.Systematic Review RegistrationOpen Science Framework, https://osf.io/sbyn3.