All articles
6 min readrunning / endurance

Second person bought 23 seconds over 10 km: how to script self-talk

At 70 km the argument in my head stops being about pace and starts being about whether to keep going. I treated that voice as weather for years. James Hardy's group at Bangor treats it as equipment: changing one word inside it raised the power output of sixteen riders over 10 km, while the ride felt the same to them 1.

Two weeks of rehearsal, 113 extra seconds.

Anthony Blanchfield's team ran the trial that put self-talk into the physiology literature. Twenty-four participants (age 24.6 ± 7.5 years, VO2max 52.3 ± 8.7 mL/kg/min) rode to exhaustion at 80% of peak power output, then spent two weeks either learning motivational self-talk or doing nothing, then rode again. The self-talk group went from 637 ± 210 to 750 ± 295 seconds, P < 0.05. The control group went from 486 ± 157 to 474 ± 169 seconds. Halfway through the ride, the self-talk group rated the effort 6.4 where it had rated 7.3 before 2. Same watts, 113 more seconds, a lower number on the effort scale.

Phillip Wallace's group repeated the design in the heat. Eighteen trained cyclists (14 men, 4 women) spent 30 minutes at 60% of peak power in 35°C and 50% humidity, then rode to exhaustion at 80%. Two weeks of self-talk training moved time to exhaustion from 487 ± 173 to 679 ± 251 seconds, P = 0.021, against 531 to 510 seconds in the control group 3.

Self-paced efforts respond too. Martin Barwood's group had 14 men ride four 10 km time trials. After the third, the motivational self-talk group's completion time fell from 1120 ± 113 to 1078 ± 96 seconds, while the group given neutral statements stayed where it was 4.

One word changed the power output.

Hardy's crossover trial is the one I find hardest to argue with. Sixteen physically active men rode 10 km twice, once using first-person self-talk and once using second-person. Second person came in at 17:25 against 17:48, p = .014, with greater power output through the trial, p = .03. Ratings of perceived exertion matched across the two rides, p = .75 1. The riders did more work without registering it as harder.

Johanne Nedergaard's group asked whether the inner voice carries any load. Two preregistered experiments, 49 and 50 participants, twelve one-minute cycling sprints each, performed alongside a verbal task, a visuospatial task or no interference. Verbal interference cost more speed than the visuospatial control in both runs: d = 0.29 against control in the first experiment, and in the second, with a harder interference task, d = 1.0 against control and d = 0.43 against the visuospatial condition 5. Occupying inner speech costs watts, which is the opposite of what you would predict if the voice were commentary.

29 runners at 60 miles gained nothing.

Alister McCormick, Carla Meijen and Samuele Marcora took the same intervention to a 60-mile overnight ultramarathon. They randomised 29 runners to motivational self-talk or a control condition. Self-talk left finishing performance unchanged, and left pre-race self-efficacy and perceived control unchanged as well. Six months later most of those runners were still using it anyway 6.

For anyone racing long, that null result carries more weight than the lab numbers above it. A 100 km race runs 10 to 15 hours and gets decided by fuelling, stomach, feet, temperature and the pacing you chose in the first two hours. A rehearsed phrase does nothing about an empty stomach at 60 km.

The meta-analysis points the same way. Antonis Hatzigeorgiadis and colleagues pooled 32 studies and 62 effect sizes for a mean of 0.48, with larger effects for fine motor tasks than gross ones, and larger effects for novel tasks than well-learned ones 7. Distance running is gross and well-learned, which puts it at the small end of that range.

Two limits are worth holding onto. All five trials above ran on a bike, in a lab, over efforts shorter than 20 minutes. And Wallace's riders finished their heat trial hotter than before the intervention, with terminating rectal temperature up from 38.5 to 38.8°C, P = 0.023 3: part of the gain came from overriding a signal that exists for a reason.

The protocol.

  1. Write four statements in the second person, each tied to a phase of the race. Mine, for 100 km: kilometre 0 to 30, "you are holding back on purpose"; 30 to 60, "you eat now, every 30 minutes"; 60 to 85, "you walk the climbs, you run the flats"; 85 to 100, "you are closing this". Second person is the one variable with a clean crossover result behind it 1.
  2. Make each statement an instruction your body can obey. "You eat now" gets obeyed or disobeyed, and you know which. "Stay strong" cannot be checked against anything.
  3. Rehearse the set for two weeks of hard sessions before you need it. The two trials that moved physiology both used a two-week learning period inside normal training 23. Nobody tested reading a list of phrases on the start line.
  4. Start using the script at the effort where your pace usually slips, rather than from the first kilometre. The lab gap opened at the halfway effort rating 2 and in the final third of a 10 km trial 4.
  5. Cap the set at four phrases, six words each. Past four I lose the thread in the last third of a long race, and that is the only test I can run on myself.
  6. Rank the script behind fuelling and pacing in your race plan. The one ultramarathon trial found nothing 6, and the pooled average for gross, well-learned tasks sits under 0.48 7. Treat it as worth a percent, not a plan.
  7. In the heat, keep the script on pace discipline. Wallace's riders ended 0.3°C hotter after the intervention 3. If your sweating drops off or your skin goes cold, the script stops and you slow down.
  8. Test it on yourself. Run the same hard session twice, two weeks apart, once with the script and once without, and record two numbers: the split of the final third, and your effort rating at halfway. Those are the two that moved in the lab 24. If the final third is no faster and the halfway rating no lower, the script is decoration for you, and the two weeks are better spent on carbohydrate tolerance.

Sources.

  1. Hardy, J., Thomas, A. V., & Blanchfield, A. W. (2019). To me, to you: how you say things matters for endurance performance. Journal of Sports Sciences, 37(18), 2122-2130. doi.org/10.1080/02640414.2019.1622240
  2. Blanchfield, A. W., Hardy, J., De Morree, H. M., Staiano, W., & Marcora, S. M. (2014). Talking yourself out of exhaustion: the effects of self-talk on endurance performance. Medicine & Science in Sports & Exercise, 46(5), 998-1007. doi.org/10.1249/MSS.0000000000000184
  3. Wallace, P. J., McKinlay, B. J., Coletta, N. A., Vlaar, J. I., Taber, M. J., Wilson, P. M., & Cheung, S. S. (2017). Effects of motivational self-talk on endurance and cognitive performance in the heat. Medicine & Science in Sports & Exercise, 49(1), 191-199. doi.org/10.1249/MSS.0000000000001087
  4. Barwood, M. J., Corbett, J., Wagstaff, C. R. D., McVeigh, D., & Thelwell, R. C. (2015). Improvement of 10-km time-trial cycling with motivational self-talk compared with neutral self-talk. International Journal of Sports Physiology and Performance, 10(2), 166-171. doi.org/10.1123/ijspp.2014-0059
  5. Nedergaard, J. S. K., Christensen, M. S., & Wallentin, M. (2023). Mind over body: interfering with the inner voice is detrimental to endurance performance. Psychology of Sport and Exercise, 68. sciencedirect.com/science/article/pii/S1469029223000961
  6. McCormick, A., Meijen, C., & Marcora, S. (2018). Effects of a motivational self-talk intervention for endurance athletes completing an ultramarathon. The Sport Psychologist, 32(1), 42-50. doi.org/10.1123/tsp.2017-0018
  7. Hatzigeorgiadis, A., Zourbanos, N., Galanis, E., & Theodorakis, Y. (2011). Self-talk and sports performance: a meta-analysis. Perspectives on Psychological Science, 6(4), 348-356. doi.org/10.1177/1745691611413136

Keep reading