The availability of task-specific feedback does not affect 20 km time trial cycling performance or test-retest reliability in trained cyclists

David N Borg; John O Osborne; Ian B Stewart; Joseph T Costello; Jonathon Headrick; Benjamin S McMaster; Samantha J Borg; Geoffrey M Minett

doi:10.1016/j.jsams.2019.12.024

The availability of task-specific feedback does not affect 20 km time trial cycling performance or test-retest reliability in trained cyclists

J Sci Med Sport. 2020 Aug;23(8):758-763. doi: 10.1016/j.jsams.2019.12.024. Epub 2019 Dec 23.

Authors

David N Borg¹, John O Osborne², Ian B Stewart², Joseph T Costello³, Jonathon Headrick⁴, Benjamin S McMaster², Samantha J Borg⁵, Geoffrey M Minett²

Affiliations

¹ Institute of Health and Biomedical Innovation, School of Exercise and Nutrition Sciences, Queensland University of Technology, Australia; The Hopkins Centre, Menzies Health Institute Queensland, Griffith University, Australia. Electronic address: david.borg@griffith.edu.au.
² Institute of Health and Biomedical Innovation, School of Exercise and Nutrition Sciences, Queensland University of Technology, Australia.
³ Extreme Environments Laboratory, Department of Sport and Exercise Science, University of Portsmouth, United Kingdom.
⁴ School of Allied Health Sciences, Griffith University, Australia.
⁵ The Hopkins Centre, Menzies Health Institute Queensland, Griffith University, Australia; Centre for Health System Reform and Integration, University of Queensland-Mater Research Institute, Australia.

PMID: 31892509
DOI: 10.1016/j.jsams.2019.12.024

Abstract

Objectives: This study examined the influence of the availability of task-specific feedback on 20 km time trial (20TT) cycling performance and test-retest reliability.

Design: Thirty trained, club-level cyclists completed two 20TT's on different days, with (feedback, FB) or without (no-feedback, NFB) task-specific feedback (i.e., power output, cadence, gear and heart rate [HR]). Elapsed distance was provided in both conditions.

Methods: During trials, ergometer variables and HR were continuously recorded, and a rating of perceived exertion (RPE) was collected every 2 km. Data were analysed using linear mixed-effects models in a Bayesian framework, and Cohen's d was calculated for standardised differences. The reliability of finish time and mean power output (PO) was determined via multiple indices, including intraclass correlations (ICC).

Results: Performance, pacing behaviour, and RPE were not statistically different between conditions. The posterior mean difference [95% credible interval] between TT1 and TT2 for FB and NFB was 10s [-5, 25] and -2s [-17, 14], respectively. In TT2, HR was statistically higher (∼8bmin^-1) in FB compared to NFB after 13 km (d = 2.08-2.25). However, this result was explained by differences in maximal HR. Finish time (FB: ICC= 0.99; NFB: ICC=0.99) and mean power output (FB: ICC=0.99; NFB: ICC=0.99) in each condition were substantially reliable.

Conclusions: The availability of task-specific information did not affect 20TT performance or reliability. Except for elapsed distance, task-specific feedback should be withheld from trained cyclists when evaluating interventions that may affect performance, to prevent participants from recalling previous performance settings.

Keywords: Behaviour; Exercise; Pacing; Reproducibility; Variation.

MeSH terms

Adaptation, Physiological
Adolescent
Adult
Athletic Performance*
Bicycling*
Competitive Behavior*
Feedback, Psychological*
Female
Humans
Male
Reproducibility of Results
Task Performance and Analysis
Young Adult