Skip to main content
NCBI home page
European Journal of Sport Science logoLink to European Journal of Sport Science
. 2024 Mar 18;24(5):577–586. doi: 10.1002/ejsc.12065

Placebo effects on kayak sprint performance in child athletes

Attila Szabo 1,2,✉
PMCID: PMC11235916

Abstract

Empirical research with adults reveals that performance‐enhancing placebo effects emerge in sports and exercise. However, there is no research on children. Coaches' messages to them could have positive, performance‐improving (placebo) or negative, performance‐impairing (nocebo) effects. This experiment examined the former by ascribing fictive performance‐enhancing properties to an ingredient of the Tic Tac mint to 12 children (aged 12.67 ± SD = 1.83 years), all elite kayakers. Another kayaker was an actor who helped enhance the credibility of the information. The children completed 2‐min kayak ergometer sprints in counterbalanced control and placebo conditions. The measures included heart rate (HR), feeling state, perceived arousal, and expected‐ and perceived‐ performances. Pre‐sprint and maximal HR during the sprint and distance completed in 2 minutes were statistically significantly higher (p < 0.005) in the placebo than in the control condition without being affected by order effects. While the expected performance in the placebo session was higher (p = 0.008), perceived performances did not differ statistically between the conditions. This study reveals a sport performance‐related objective, but not subjective, placebo effect in children after a short kayak sprint. The findings have practical implications for performance‐related messages children receive from their coaches and others and show how their altered beliefs can influence their performance.

Keywords: beliefs, children, conditioning, exercise, sports, training

Highlights

  • This study is the first to examine placebo effects in elite child athletes.

  • A placebo effect occurred on child kayakers' 2‐min sprint performance.

  • In this study, objective but not subjective placebo effects emerged.

1. INTRODUCTION

The placebo effect is a desirable (pleasing) outcome in response to a believed‐to‐be‐beneficial agent, treatment, or event, while the nocebo effect is the opposite (Hurst et al., 2019). Placebos are primarily used in medicine and pharmacology to test the net effects of new drugs beyond the healing effects associated with expectations and no treatment at all. However, recent sports and exercise science research suggests that placebo and nocebo effects, through very complex neural mechanisms, can also affect sports performance (Hurst et al., 2019). Beedie et al. (2018) summarized the current knowledge in this area in four points: (1) placebo intervention studies generally reveal a placebo effect in contrast to baseline or no‐treatment conditions; (2) placebo effects could benefit athletic performance; (3) placebo effects might account for a substantial proportion of the overall treatment effect; and (4) placebo effects could contribute significantly to exercise‐related outcomes, accounting for up to 50% of both subjective (perceived) and objective (actual) measures) effects. Due to this journal's word limit and space required to describe the research methods, past papers on placebo effects in sports are not reviewed here. Instead, the reader is referred to review articles and theoretical papers on this topic (Beedie et al., 2018; Bérdi et al., 2011; Hurst et al., 2019; Szabo, 2013, 2023).

Until now, sports and exercise science studies investigating placebo effects tested adults only (Hurst et al., 2019). Similar research on children's sports performance is lacking, possibly due to limited access to this sample, coaches' and parents’ resistance to experimenting with child athletes, and ethical considerations. The last relates primarily to the deception part of placebo research, a sensitive issue in research with children (Fisher, 2005).

According to the American Psychological Association's (APA's) Ethical Principles of Psychologists and Code of Conduct (APA, 2017), researchers using deceptive methodologies have an ethical responsibility to determine (1) whether deception is justified, (2) does not cause pain or emotional harm, and (3) explain the reason for deception to the participants at the latest after data collection (Standard 8.07a, Deception in Research). Therefore, while subject to conditions, this form of research might be justified if alternative methods do not permit answering a specific research question and APA's Standard 8.07a is respected.

Fisher (2005) proposed four guidelines to be considered when deception is used in experiments with children. They: (1) evaluate the scientific validity and value of deceptive research in contrast to alternative methodologies, (2) avoid or minimize experimental risk and potential harm, (3) in addition to the child's consent, obtain parental or guardian consent, and (4) debrief at the level of understanding of the child. Studies suggest that children as young as nine can express their preferences for research participation based on their comprehension of research objectives, procedures, risks, and confidentiality (Fisher, 2005). They also seem to understand that their involvement can benefit others (Fisher, 2005).

Placebo research without deception or concealment of information is only possible in an open‐label placebo (OLP) design (Kaptchuk, 2018), for which young children may not be suitable based on the complex neural mechanisms described by Kaptchuk. Even with adults, Saito et al. (2020) found only two studies in sports science, which yielded incongruent results. Therefore, at this time, OLP with young child athletes is unfounded, but it may be feasible with more mature adolescents. Still, based on the primarily positive findings of 32 studies in sports and exercise with adults, reviewed by Hurst et al. (2019), placebo intervention using deception might benefit children, too. However, what could be the benefits of such research?

The rationale for this research may be relatively simple. For example, placebo effects in children are substantially greater than in adults, based on reviews of clinical studies (Weimer et al., 2013). The negative relationship between age and placebo effects was more recently confirmed in a non‐clinical experimental setting, too, showing that placebo hypoalgesia was especially evident in children under 14 years (Gniß et al., 2020). The positive effects occurred via both expectation manipulation and conditioning interventions. Such findings prompted the authors to conclude that “These results encourage the use of placebo effect in clinical practice, particularly for younger children (Gniß et al., 2020, p. 1191).” Since in other than sports or exercise domains, children are more responsive to placebos than adults, it may be expected that these findings could be replicated in sports performance settings and the minor effects of fictive performance enhancers seen in adults (Hurst et al., 2019) might be more significant in children. The emerging implication is that placebo interventions‐induced benefits on performance (if any) are positively related to physical self‐concept and intrinsic motivation (Lohbeck et al., 2021), thus benefiting children's sports adherence and positive psychological experiences.

On ethical grounds, first, parents deceive their children with good intentions (Safitri, 2020). For example, the stories about Santa Claus, the Easter Bunny, and the Tooth Fairy are deceptive but are presumed to make children happy (Blair et al., 1980). A retrospective study with university students suggested that children's attitudes toward white lies are generally positive because of good intentions behind them that could benefit the receiver (Cargill & Curtis, 2017).

Second, about half of the coaches, especially those working with elite athletes (>60%), use some placebo in their coaching practice to provide a mental ergogenic aid to their athletes (Szabo & Müller, 2016). While ethically debatable, using a harmless placebo that improves performance could teach the child how their mind power could aid them in performing better. Therefore, the treatment could increase self‐efficacy, persistence, and achievement, based on Schunk's (1995) self‐efficacy theory. In this sense, using a harmless placebo is done with good intentions to benefit the receiver (Szabo, 2023) and can be a crutch for those stalled at a certain level of performance. The research question posed in this experimental study is whether a placebo intervention could benefit elite child athletes' sports performance.

This study tested elite child kayakers using belief (expectancy) manipulation through a fictive story associated with the Tic Tac mint to induce a placebo effect. The story was told by the coach and reinforced by an actor. Based on Hurst et al.’s (2019) review, the hypothesis was that the placebo intervention would enhance performance. As this is the first study to examine placebo effects in child athletes, exploring how children feel during such an intervention was necessary. Therefore, feeling states and felt arousal were assessed to determine the core affect (Russell, 2003) based on affective valence and activation in various test situations. Finally, expected‐ and perceived performances were assessed to examine the placebo's subjective effects. Their relationship to performance was expected to be positive.

The practical significance of this research, beyond determining the possible benefits of the placebo effects on child athletes' sports performance, is that it could help coaches reveal to children the power of their minds. Then, being convinced about the impact of their beliefs on athletic performance, child athletes can voluntarily manipulate their thoughts through rituals, self‐talk, or suggestions, all being associated with better athletic performance. Indeed, Beedie et al. (2015) draw an analogy between the placebo effects and sports psychology interventions aimed at modifying beliefs (i.e., to raise self‐confidence or alleviate anxiety) since both aim to build a mental bridge leading from what is currently being achieved to what is achievable.

2. MATERIALS AND METHODS

2.1. Participants

2.1.1. Ethics

The Research Ethics Board (REB) of the Faculty of Education and Psychology at Eötvös Loránd University granted ethical clearance for the study (No. 2022/441). Further, the REB ensured that the work conforms to the ethical guidelines of the British Psychological Society Code of Human Research Ethics (British Psychological Society, 2021). Additionally, the protocol followed the research principles with human participants of the Helsinki Declaration (World Medical Association, 2013). Finally, all participants and their parents/guardians read and signed an informed consent form before participating in the study.

2.1.2. Recruitment and sample characteristics

The manager of a kayak sporting club located in Vác, Hungay granted permission for the study with the club's young athletes subject to parents'/guardians' consent. After his approval and ethical clearance from the REB, the children were asked to participate in the study if their legal guardians supported that. In total, 13 athletes agreed to participate with their parents/guardians’ consent.

One was a 17‐year‐old boy assuming the role of an actor in the study. He knew the aims of the research. However, he vowed confidentiality until the completion of the study. His task was to tell the rest of the participants—before the study—that he had tried the placebo agent (described later), and despite not believing it, his 2‐min kayak sprint performance increased by 30% compared to his best previous performance.

The remaining 12 participants (mean age = 12.67 ± SD = 1.83, range 10–16 years) were primarily boys except for one girl (aged 10 years). Participants have practiced kayaking for at least 2 years and have had several successful competition seasons. The sample comprised athletes achieving second to ninth place in Hungarian national championships. Further, all of them were among the top 20 kayakers in the country within their age group. The young athletes have trained six times a week for over 2 hours since they were eight. Their health was excellent based on the regular and mandatory medical examinations. Their coach, trained in research, helped in collecting data and delivering the placebo message.

2.2. Materials

2.2.1. Kayak ergometer

A ‘KayakPro Speedstroke’ (Infosports Co. Ltd.) kayak ergometer was used to measure the distance covered in meters and the stroke rate over 2 minutes of kayak sprint. This ergometer has an adjustable shaft to adjust paddling resistance and yield a user‐specific kayak feel. Additionally, it has an easy‐adjust, quick‐release footrest, an adjustable height and angle seat, and a large console display.

2.2.2. Heart rate monitor

Heart rate (HR) measures were obtained with a ‘Polar RS400’ (Polar Electro Oy) wearable HR monitor connected to a Polar WearLink® 31 chest belt. This instrument has good criterion‐related validity and test‐retest reliability, and it is suitable for HR monitoring in exercise settings (Engström et al., 2012).

2.2.3. The placebo agent

The placebo agent was a commercially available ordinary Tic Tac mint manufactured by Ferrero, Italy. Specifically, a fictive discovery linked to one of its components listed on its ingredients label, the Gum Arabic (a thickener compound), was claimed to be a performance enhancer. Therefore, while the physical agent was the Tic Tac, the actual placebo was the belief manipulation associated with its ingredient. This ingredient is a gummy exudation that hardens in the air and flows naturally or via cuts made in tree trunks and branches of the L. Acacia Senegal, L. Willdenow, and several other African Acacia species. It is considered a dietary fiber compatible with the human diet having several health benefits (Phillips & Phillips, 2011). Its picture (Wikimedia Commons, 2023) was displayed for visual association. The coach told them that new research shows this ingredient is a potent performance enhancer, but he is skeptical about such effects. Nevertheless, a kayaker in the club tried it, and unbelievably he beat his record on a 2‐min sprint by 30%. Then, an actor kayaker confirmed this statement.

2.2.4. Psychological measures

The Feeling Scale (FS; Hardy & Rejeski, 1989) assessed the feeling states (affective valence) before and after kayak sprints. This scale is rated from −5 to +5, reflecting feelings ranging from very bad to very good. The Felt Arousal Scale (FAS; Svebak & Murgatroyd, 1985), which is rated on a 6‐point Likert scale ranging from 1 (very low arousal) to 6 (very high arousal), gauged the subjectively perceived arousal before and after the kayak sprints. Both scales have good validity based on the reports of the developers. The results of these two single‐item scales determine the core affect. Core affect is a conscious, general, non‐nreflective psychophysiological feeling state, such as good or bad, tired, or energetic (Russell, 2003).

To determine the expected performance before the 2‐min kayak sprints in two conditions, participants had to respond to the questions: “What do you think, how will you perform during this trial?” on a Likert scale ranging from 1 (very poorly) to 10 (very good). Similarly, to assess their perception of the actual performance after the 2‐min sprint, they had to respond to the question: “How did you perform on this trial?” on a Likert scale ranging from 1 (very poor) to 10 (very good).

2.3. Procedure

2.3.1. Placebo delivery

Figure 1 displays the experimental protocol. Initially, participants had a group meeting. During this meeting, the coach told them he read a scientific paper about a novel discovery that the ordinary Tic Tac mint contains a harmless ingredient, Gum Arabic, a powerful performance enhancer. He then displayed a large picture of the ingredients label on the Tic Tac box listing the Gum Arabic, followed by the plant itself (Wikimedia Commons, 2023). Furthermore, he told children that the Tic Tac would probably be banned before sports competitions once the World Anti‐Doping Agency (WADA) becomes aware of its sports performance‐enhancing ingredient. He then elaborated further by stressing that the research article reported that one mint is as efficient as 10 or more, so the effect is not dose dependent.

FIGURE 1.

FIGURE 1

Open in a new tab

The experimental protocol.

Finally, the coach asked the children if they would also like to experiment rigorously (under his supervision using precise measures readable on the ergometer) with the harmless Tic Tac mint. As expected, some children raised their hands immediately, followed by all within seconds. To maximize the expectancy associated with the Tic Tac‐induced performance increase, the coach called in the older acting kayaker, who told participants that the coach assessed his performance in a 2‐min ergometer sprint after ingesting a Tic Tac. He continued, “I thought this was a joke, but when I looked at my sprint time being 30% better than my best time on this distance, I was astonished.”

2.3.2. Experimental protocol

Participants were randomized and tested in counterbalanced order. The research occurred in the 2022–2023 winter in a medium‐sized room (14 m2) serving as a test facility within the participants' training center. Familiarity with the research environment eliminates or reduces potential anxiety caused by novelty and unfamiliar settings, which is critical in testing children (Lerwick, 2016). The room contained the kayak ergometer, a small work desk, an office chair, and an armchair. Upon entering the test environment in a well‐lit (4000 lumens), humidity‐ (60%) and temperature‐controlled (20°C) room, participants received verbal explanations about the experimental procedure. If they had any questions, the coach answered them. Subsequently, they were instructed to relax in an armchair while reading and signing the consent form their parent or guardian had already signed.

All experiments occurred during the daytime; one test lasted about 15–20 min. It started with mounting the HR monitor on the participant's wrist and the signal‐transmitting belt on the chest. After 3 minutes of rest, anticipatory heart rate (AHR), feeling state, and felt arousal were recorded. In the experimental condition, the participants ingested a Tic Tac mint and then indicated their expectancy for the upcoming 2‐min kayak sprint performance. (The same sequence occurred in the control condition, except for ingesting the mint that was replaced with the instruction to try to beat their previous record). Then they sat on the ergometer and set themselves ready for the sprint. Ergometer resistance matched the resistance used in training (lower for younger participants and higher for older ones but held constant for the same person in the control and experimental trial). The ergometer's display was visible only to the coach, who recorded the distance traveled in 2 minutes at the end of the sprint and the number of strokes. He also recorded the maximal heart rate (MHR) during the sprint. Immediately after the sprint, the participant indicated the perceived performance on the 10‐point Likert scale. Subsequently, after 3 minutes, a recovery HR was recorded (RHR).

The participant then went for a two‐hour rest in another room, watched documentaries about techniques in kayaking, and waited to be retested. If the first in the sequence was the experimental session, participants were told that their performance was outstanding and better than their earlier sprint distances. On the other hand, if they underwent the control session first, they were told that their performance was like what they produced during regular training. Still, they could achieve a longer distance after ingesting the Tic Tac in the subsequent trial. Since testing was individual, those who finished earlier could talk to (or text on mobile) others still waiting to be tested. Therefore, all participants were told they performed between 30% and 40% better with the Tic Tac.

After completing the study, a meeting with the participants and their parents was called immediately. Its scope was to debrief them, provide the study's rationale, and explain the need for deception to induce altered beliefs. At this stage, participants and their parents had a chance to request the deletion of their coded (anonymously recorded) data. None of them did. Finally, the participants and their parents were thanked for their collaboration.

2.4. Data reduction and analyses

Data were transcribed into an Excel file, verified, and imported into a Statistical Package for Social Sciences (SPSS; IBM Corp, 2021) file for statistical analyses. Given the small sample size, nonparametric tests were the most appropriate (Fagerland, 2012) for analyzing the data. These tests are less powerful and carry the risk of Type II errors, but they robustly guard against Type I errors. Additionally, the statistical significance levels (alpha) were corrected in all instances of multiple tests. First, the possible order effects were examined with the independent samples Mann‐Whitney U tests. Second, differences between the control and placebo conditions were tested with the ‘related‐samples Wilcoxon signed‐ranked tests.’ Furthermore, the patterns in core affect—before and after the control and experimental sessions—were visually examined based on the affective‐valence‐activation axes of the circumplex model (Ekkekakis & Petruzzello, 2002). Finally, Spearman's rank correlations were used to test the relationships between expected and perceived performance and the distances completed in the two sessions.

3. RESULTS

The test of the order effects (see Table 1) did not result in statistically significant differences. Furthermore, despite the results obtained on the feeling states before the placebo and after the control conditions yielded results that exceeded the adopted statistical significance level (i.e., alpha [α] = 0.05), after correcting the alpha for multiple tests, these results were no longer statistically significant.

TABLE 1.

Independent samples Mann‐Whitney U test results for order effects (i.e., those exposed to placebo in the first or the second session).

Measures U Z p RBC a
Age 15.00 −0.491 0.624 −0.167
AHR control 16.00 −0.320 0.749 −0.111
MHR control 18.00 0.000 1.000 0.000
RHR control 12.50 −0.882 0.378 0.306
Feeling before control 12.00 −0.983 0.325 −0.333
Feeling after control 3.00 −2.441 0.015 b −0.833
Arousal before control 13.50 −0.742 0.458 0.250
Arousal after control 17.00 −0.166 0.868 0.056
Expectancy control 10.50 −1.345 0.179 −0.417
Perceived performance control 14.00 −0.654 0.513 −0.222
Distance in meters control 10.00 −1.281 0.200 −0.444
Total strokes control 16.50 −0.241 0.809 −0.083
AHR placebo 17.00 −0.162 0.871 −0.056
MHR placebo 16.00 −0.321 0.748 0.111
RHR placebo 15.00 −0.480 0.631 0.167
Feeling before placebo 5.00 −2.111 0.035 b −0.722
Feeling after placebo 6.50 −1.920 0.055 −0.639
Arousal before placebo 9.50 −1.422 0.155 −0.472
Arousal after placebo 17.50 −0.090 0.929 0.028
Expectancy placebo 11.50 −1.090 0.276 −0.361
Perceived performance placebo 9.00 −1.475 0.140 −0.500
Distance in meters placebo 12.00 −0.961 0.337 −0.333
Total strokes placebo 15.00 −0.481 0.630 −0.167
Open in a new tab

Abbreviations: AHR, anticipation heart rate; MHR, MHR during the sprint; RHR, recovery heart rate.

a

RBC, Rank Biserial Correlation reflecting the effect size.

b

Statistically not significant after the correction of the alpha (new α = .002) with the Bonferroni method.

Examining the differences between the placebo and control conditions (Table 2) yielded three statistically significant results after the Bonferroni correction of the alpha. These results emerged in AHR, MHR, and distance in meters completed in the two‐minute sprint. The power (1 — β) achieved for these tests ranged between 0.51 and 0.99 as calculated from effect size, sample size, and α level for Wilcoxon Signed Ranked tests using the G*Power (v. 3.1.9.7) software (Faul et al., 2007). In addition, there was a noteworthy trend in expectation, higher in the placebo than control session that approached (0.008) but did not reach the corrected level (0.005) of statistical significance. Given that (1) these tests were pre‐planned, (2) nonparametric analyses have low statistical power, and (3) the Bonferroni correction method decreases the statistical power further (Armstrong, 2014), the results in expectancy, despite only emerging as a trend, should be considered conceptually meaningful. This argument is especially valid based on the effect size for expectancy in Table 2 (r = 0.770, converted to Cohen's d = 2.41), which is large (Rice & Harris, 2005), and the fact that this test was planned as a manipulation check.

TABLE 2.

Comparison between the control and placebo sessions as calculated with related samples Wilcoxon sign‐ranked test, also showing the means and standard deviations.

Measures Mean ± SD control Mean ± SD placebo D 1 from control W z p r a
AHR 63.58, 8.08 69.67, 4.70 9.57% 2.00 −2.903 0.004 c −0.837
MHR 178.25, 20.47 188.92, 13.54 5.99% 2.00 −2.903 0.004 c −0.837
RHR 108.08, 20.31 113.17, 18.57 4.71% 19.00 −1.569 0.125 −0.453
Feeling before 2.00, 1.65 2.75, 1.77 37.50% 9.00 −1.260 0.222 −0.364
Feeling after 2.42, 1.83 3.00, 1.60 23.97% 0.00 −1.604 0.181 −0.463
Arousal before 4.17, 1.59 4.67, 1.16 11.99% 18.00 −0.968 0.347 −0.279
Arousal after 4.50, 1.68 5.25, 1.06 16.65% 8.00 −1.718 0.081 −0.496
Expectation 6.75, 1.06 7.92, 1.38 17.33% 0.00 −2.666 0.008 b −0.770
Perceived performance 7.25, 2.18 8.17, 1.70 12.69% 9.00 −1.599 0.121 −0.462
Distance in meters 341.47, 34.75 353.93, 40.09 3.65% 3.00 −2.824 0.002 c −0.815
Total strokes 96.00, 8.77 98.58, 15.88 2.69% 27.00 −0.533 0.623 −0.154
Open in a new tab

Note: D 1 = Mean difference (in percent) compared to the control session.

a

r = effect size for Wilcoxon sign‐ranked test (r = Z/√N) based on Rosenthal (1994).

b

To be considered as a ‘trend’ because statistically, it is not significant after the correction of the alpha (new α = .005) with the Bonferroni method.

c

Statistically significant after Bonferroni correction of the alpha.

The plotting of the intersection of feeling states and felt arousal before and after placebo and control sessions, based on the circumplex model of affect showed a continuous upward switch in core affect. As seen in Figure 2, core affect was lowest before the control session and highest after the placebo section. However, in all instances, it was in the high activation‐positive valence quadrant of the model.

FIGURE 2.

FIGURE 2

Open in a new tab

Core affect, as represented by the intersection of feeling state and felt arousal, before the control session (A), after the control session (B) before the placebo session (C) and after the placebo session (D).

The expected and perceived performance correlated statistically significantly in both control (rho [ρ] = 0.840) and placebo (ρ = 0.738) conditions. Subsequently, a standard Fisher's z‐transformation of the Spearman coefficients (Myers & Sirois, 2004) yielded no statistical difference between these correlations (z = 0.58, p = 0.562). Further, the expected and perceived performance did not correlate with the distance completed in either test condition. Still, the correlation between the completed distances in the two test sessions was high (ρ = 0.937).

4. DISCUSSION

This experiment demonstrated that children's kayak sprint performance can be enhanced via belief manipulation, associated with an ordinary and harmless mint. The effect size of the difference between the control and placebo session was large, far exceeding the medium effect sizes reported for adults in literature reviews on placebo effects in sports and exercise. For example, an earlier meta‐analysis of 14 studies (Bérdi et al., 2011) reported an overall medium effect size (d = 0.40). Similarly, a more recent systematic review of 32 studies (Hurst et al., 2019) found a moderate effect size (d = 0.38). In contrast, the effect size in the current study was much larger. For example, if converted to Cohen's d, the effect size r = ‐. 815 (from Table 2) is equivalent to d = −2.81 (Fritz et al., 2012), more than seven folds higher than the average effect sizes reported in the reviews of placebo research in sports conducted with adults.

Noteworthy is that five statistically non‐significant results (Table 2; excluding arousal before sprints and total strokes) yielded medium (i.e., r > 0.3; Bhandari, 2023) effect sizes that might have emerged as statistically significant with a larger sample size. Indeed, since only 12 young athletes were studied, the statistical power was low and Type II errors might have masked some treatment effects. However, experimental studies on placebo effects in sports often must rely on small sample sizes as evidenced in the review article of Hurst et al. (2019) in which 9/32, or close to one third of the reviewed studies employed identical to or smaller samples than the current work.

The observed large effect size associated with performance might be due to at least two factors. One is that in addition to the trusted coach's fictive story, the grass‐like picture of the plant (Wikimedia Commons, 2023) and the acting kayaker—claiming to have benefited from it—could have substantially enhanced the children's belief/expectancy about the ergogenic aid of the Gum Arabic in the Tic Tac. Another is that young children are quick to act on the information provided by adults, and their trust is higher/highest in those close to them (Harris & Corriveau, 2011), like their coach and fellow kayaker.

Indeed, their trust in the placebo agent was confirmed by the higher AHR and MHR, showing that they both anticipated better performance and worked harder in the placebo than in the control session. This argument is supported by the higher expectancy in the placebo session, which despite emerging only as a trend, was associated with a large effect size (see Table 2). Additionally, the visual examination of the core affect (Figure 2) shows that children felt slightly better before and after the placebo session than in the control session. However, the core affect was in the pleasant valence‐high activation quadrant of the circumplex model throughout the study, suggesting that children felt good during the experiment. Since feeling states and arousal before and after placebo and control sessions did not differ statistically significantly, the visually noticeable upward shift remains only a trend that could emerge to be significant in a study with a larger sample size, which requires further experimental scrutiny.

Noteworthy is that, while MHR was 5.99% greater in the placebo session, the total stroke number, despite being 2.69% higher in the placebo session, did not differ statistically in the two conditions, suggesting that the children might have pooled stronger and achieved longer rather than more strokes. However, it is also possible that statistical differences did not emerge because of lack of power (Type II error). Despite the harder work, they did not see their performance significantly differently in the two experimental conditions based on statistical tests. Yet, on the average, they perceived their performance in the placebo session 12.69% better than that in the control condition, which was accompanied by a medium, close to large, effect size, suggesting that subjectively perceived placebo effects might have been missed due to Type II error. Still, it is possible that the objective placebo effects are not paired with subjective effects. In such case, the findings agree with a recent work showing that regular Kinesio Tape users manifested higher grip strength with Kinesio Tape in contrast to the tapeless condition, that was attributed to a conditioned placebo effect, while non‐users showed identical grip strength with and without Kinesio Tape, but the perceived performance did not differ in any of the groups (Mak et al., 2019).

In this study, there was no correlation between actual and perceived performance or expectations, but the latter two were strongly related in both conditions. These results agree with a recent finding that placebo pill‐induced expectation, associated with postural stability, predicted the perceived performance (Russell et al., 2022). In a similar experiment, Horváth et al. (2022) came to the same conclusion and conjectured that there is dissociation between actual (objective) and perceived (subjective) performance. While different placebos may work differently and children respond differently than adults (Weimer et al., 2013), it is possible that expectations work at more conscious levels while actual placebo effects could be more subconscious (Barrett et al., 2006).

In applied settings, placebo effects in athletic performance gained via good‐intentioned deception and divulged to young athletes is a proof for the power of the mind that could later be used by the athlete—through rituals, self‐talk, or auto‐suggestions—to maximize the sports performance. Furthermore, while research on OLP in sports is still in its infancy, such research with child athletes who are mature enough to self‐administer an OLP may yield results that could boost the psychological aspects of their sports performance.

4.1. Limitations

Some limitations of the current research call for caution while interpreting and generalizing the findings. The first is that the elite child kayakers studied may not be representative of children involved in other sports or at different levels of competition. The second is that, while other studies with smaller sample sizes were published in this area, the relatively small sample prompted the use of nonparametric tests with low statistical power, which was further decreased by Bonferroni correction. While these methods have guarded robustly against Type I error, the likelihood of Type II error was high. Therefore, future studies should replicate the current findings with a larger sample size.

5. CONCLUSIONS

This study shows for the first time that placebo effects influence a sport performance in elite child athletes. A 3.65% increase in distance accomplished during a 2‐min kayak sprint is comparable to the average percentage changes (3.98%) stemming from 67 different placebo interventions, reported in 32 research papers, with adults (Hurst et al., 2019). Expectancy and anticipatory heart rate were statistically significantly higher in the placebo than in the control condition. Furthermore, expectancy was positively related to perceived (subjective) but not the actual (objective) performance. The former did not differ statistically between the two test conditions, but it was 12.69% higher in the placebo than in the control session. It is possible that some treatment effects were obscured due to lack of power yielding Type II error, as based on moderate to large effects sizes in several measures. Therefore, further examination of placebo effects in children's sport is warranted, possibly using a larger sample size, which, nevertheless, is difficult due to ethical, parental, and sport management concerns. It is unknown the extent to which coaches use placebos with children, but this work suggests that it might work and could prove to children how their mind could affect objective athletic performance. Still, using placebos with children is subject to robust ethical scrutiny.

CONFLICT OF INTEREST STATEMENT

The author has no conflict of interest to declare.

ACKNOWLEDGMENTS

Many thanks to Mr. Dominic K is, the participating children's coach, who assisted in participant recruitment and data collection. The author is extremely grateful to two anonymous reviewers for their constructive reviews that helped to improve the presentation of the report. The author did not receive funding for this study.

REFERENCES

  1. American Psychological Association . 2017. Ethical Principles of Psychologists and Code of Conduct Effective Date June 1, 2003, with Amendments Effective June 1, 2010, and January 1, 2017. http://www.apa.org/ethics/code/index.html.
  2. Armstrong, Richard A. 2014. “When to Use the Bonferroni Correction.” Ophthalmic and Physiological Optics 34(5): 502–508. 10.1111/opo.12131. [DOI] [PubMed] [Google Scholar]
  3. Barrett, Bruce , Muller Daniel, Rakel David, Rabago David, Marchand Lucille, and Scheder Joanne Caroline. 2006. “Placebo, Meaning, and Health.” Perspectives in Biology and Medicine 49(2): 178–198. 10.1353/pbm.2006.0019. [DOI] [PubMed] [Google Scholar]
  4. Beedie, Christopher , Benedetti Fabrizio, Barbiani Diletta, Camerone Eleanora, Cohen Emma, Coleman Damian, Davis Arran, et al. 2018. “Consensus Statement on Placebo Effects in Sports and Exercise: The Need for Conceptual Clarity, Methodological Rigour, and the Elucidation of Neurobiological Mechanisms.” European Journal of Sport Science 18(10): 1383–1389. 10.1080/17461391.2018.1496144. [DOI] [PubMed] [Google Scholar]
  5. Beedie, Chris , Foad Abby, and Hurst Philip. 2015. “Capitalizing on the Placebo Component of Treatments.” Current Sports Medicine Reports 14(4): 284–287. 10.1249/jsr.0000000000000172. [DOI] [PubMed] [Google Scholar]
  6. Bérdi, Márk , Köteles Ferenc, Szabó Attila, and Bárdos György. 2011. “Placebo Effects in Sport and Exercise: A Meta‐Analysis.” European Journal of Mental Health 6(2): 196–212. 10.5708/ejmh.6.2011.2.5. [DOI] [Google Scholar]
  7. Bhandari, P. 2023. What Is Effect Size and Why Does it Matter? (Examples). Scribbr: Retrieved from.https://www.scribbr.co.uk/stats/effect‐sizes/. [Google Scholar]
  8. Blair, John R. , McKee Judy S., and Jernigan Louise F.. 1980. “Children's Belief in Santa Claus, Easter Bunny and Tooth Fairy.” Psychological Reports 46(3): 691–694. 10.2466/pr0.1980.46.3.691. [DOI] [Google Scholar]
  9. British Psychological Society, BPS . 2021. Code of Human Research Ethics. https://www.bps.org.uk/guideline/bps‐code‐human‐research‐ethics.
  10. Cargill, Joscelyn R. , and Curtis Drew A.. 2017. “Parental Deception: Perceived Effects on Parent‐Child Relationships.” Journal of Relationships Research 8: 1–8. 10.1017/jrr.2017.1. [DOI] [Google Scholar]
  11. Ekkekakis, Panteleimon , and Petruzzello Steven J.. 2002. “Analysis of the Affect Measurement Conundrum in Exercise Psychology: IV. A Conceptual Case for the Affect Circumplex.” Psychology of Sport and Exercise 3(1): 35–63. 10.1016/s1469-0292(01)00028-0. [DOI] [Google Scholar]
  12. Engström, Elise , Ottosson Elin, Wohlfart Björn, Grundström Nils, and Wisén Anita. 2012. “Comparison of Heart Rate Measured by Polar RS400 and ECG, Validity and Repeatability.” Advances in Physiotherapy 14(3): 115–122. 10.3109/14038196.2012.694118. [DOI] [Google Scholar]
  13. Fagerland, Morten W. 2012. “t‐Tests, Nonparametric Tests, and Large Studies—A Paradox of Statistical Practice?” BMC Medical Research Methodology 12(1): 1–7. 10.1186/1471-2288-12-78. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Faul, Franz , Erdfelder Edgar, Lang A.‐Georg, and Buchner Axel. 2007. “G*Power 3: A Flexible Statistical Power Analysis Program for the Social, Behavioral, and Biomedical Sciences.” Behavior Research Methods 39(2): 175–191. 10.3758/BF03193146. [DOI] [PubMed] [Google Scholar]
  15. Fisher, Celia B. 2005. “Deception Research Involving Children: Ethical Practices and Paradoxes.” Ethics & Behavior 15(3): 271–287. 10.1207/s15327019eb1503_7. [DOI] [PubMed] [Google Scholar]
  16. Fritz, Catherine O. , Morris Peter E., and Richler Jennifer J.. 2012. “Effect Size Estimates: Current Use, Calculations, and Interpretation.” Journal of Experimental Psychology: General 141(1): 2–18. 10.1037/a0024338. [DOI] [PubMed] [Google Scholar]
  17. Gniß, Silke , Kappesser Judith, and Hermann Christiane. 2020. “Placebo Effect in Children: the Role of Expectation and Learning.” Pain 161(6): 1191–1201. 10.1097/j.pain.0000000000001811. [DOI] [PubMed] [Google Scholar]
  18. Hardy, Charles J. , and Rejeski W. Jack. 1989. “Not What, but How One Feels: The Measurement of Affect during Exercise.” Journal of Sport & Exercise Psychology 11(3): 304–317. 10.1123/jsep.11.3.304. [DOI] [Google Scholar]
  19. Harris, Paul L. , and Corriveau Kathleen H.. 2011. “Young Children’s Selective Trust in Informants.” Philosophical Transactions of the Royal Society B 366(1567): 1179–1187. 10.1098/rstb.2010.0321. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Horváth, Áron , Szabo Attila, Gál Vera, Suhaj Csilla, Aranyosy Blanka, and Köteles Ferenc. 2022. “Expectations Predict Perceived but Not the Actual Performance in Postural Stability.” SSRN Electronic Journal. 10.2139/ssrn.4236818. [DOI] [Google Scholar]
  21. Hurst, Philip , Schipof‐Godart Lieke, Szabo Attila, Raglin John, Hettinga Florentina, Roelands Bart, Lane Andrew, Foad Abby, Coleman Damian, and Beedie Chris. 2019. “The Placebo and Nocebo Effect on Sports Performance: A Systematic Review.” European Journal of Sport Science 20(3): 279–292. 10.1080/17461391.2019.1655098. [DOI] [PubMed] [Google Scholar]
  22. IBM Corp . 2021. Released 2021. IBM SPSS Statistics for Windows, Version 28.0. Armonk, NY: IBM Corp. [Google Scholar]
  23. Kaptchuk, Ted J. 2018. “Open‐Label Placebo: Reflections on a Research Agenda.” Perspectives in Biology and Medicine 61(3): 311–334. 10.1353/pbm.2018.0045. [DOI] [PubMed] [Google Scholar]
  24. Lerwick, Julie L. 2016. “Minimizing Pediatric Healthcare‐Induced Anxiety and Trauma.” World Journal of Clinical Pediatrics 5(2): 143. 10.5409/wjcp.v5.i2.143. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Lohbeck, Annette , von Keitz Philipp, Hohmann Andreas, and Daseking Monika. 2021. “Children’s Physical Self‐Concept, Motivation, and Physical Performance: Does Physical Self‐Concept or Motivation Play a Mediating Role?” Frontiers in Psychology 12: Article 669936. 10.3389/fpsyg.2021.669936. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Mak, Dominic N.‐Tung , Au Ivan P.‐Hung, Chan Mavis, Chan Zoe Y.‐Shan, An Winko Wenkang, Zhang Janet Hanwen, Draper David, and Cheung Roy T.‐Hei. 2019. “Placebo Effect of Facilitatory Kinesio Tape on Muscle Activity and Muscle Strength.” Physiotherapy Theory and Practice 35(2): 157–162. 10.1080/09593985.2018.1441936. [DOI] [PubMed] [Google Scholar]
  27. Myers, Leann , and Sirois Maria J.. 2004. “Spearman Correlation Coefficients, Differences between.” Encyclopedia of Statistical Sciences 12: 1–3. 10.1002/0471667196.ess5050. [DOI] [Google Scholar]
  28. Phillips, Aled O. , and Phillips Glyn O.. 2011. “Biofunctional Behaviour and Health Benefits of a Specific Gum Arabic.” Food Hydrocolloids 25(2): 165–169. 10.1016/j.foodhyd.2010.03.012. [DOI] [Google Scholar]
  29. Rice, Marnie E. , and Harris Grant T.. 2005. “Comparing Effect Sizes in Follow‐Up Studies: ROC Area, Cohen's D, and R.” Law and Human Behavior 29(5): 615–620. 10.1007/s10979-005-6832-7. [DOI] [PubMed] [Google Scholar]
  30. Rosenthal, R. 1994. “Parametric Measures of Effect Size.” In The Handbook of Research Synthesis, edited by Cooper H. and Hedges L. V., 231–244. New York: Russell Sage Foundation. [Google Scholar]
  31. Russell, James A. 2003. “Core Affect and the Psychological Construction of Emotion.” Psychological Review 110(1): 145–172. 10.1037/0033-295X.110.1.145. [DOI] [PubMed] [Google Scholar]
  32. Russell, Katherine , Duncan Michael, Price Michael, Mosewich Amber, Ellmers Toby, and Hill Mathew. 2022. “A Comparison of Placebo and Nocebo Effects on Objective and Subjective Postural Stability: a Double‐Edged Sword?” Frontiers in Human Neuroscience 16. 10.3389/fnhum.2022.967722. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Safitri, Rika Nurismah . 2020. “Exploring Reasons Why Parents Use White Lies to Golden Age Children in Al Ghazaly Kindergarten.” Abqari Journal 23(1): 145–155. 10.33102/abqari.vol23no1.248. [DOI] [Google Scholar]
  34. Saito, Tiemi , Barreto Gabriel, Saunders Bryan, and Gualano Bruno. 2020. “Is Open‐Label Placebo a New Ergogenic Aid? A Commentary on Existing Studies and Guidelines for Future Research.” Sports Medicine 50(7): 1225–1229. 10.1007/s40279-020-01285-w. [DOI] [PubMed] [Google Scholar]
  35. Schunk, Dale H. 1995. “Self‐efficacy, Motivation, and Performance.” Journal of Applied Sport Psychology 7(2): 112–137. 10.1080/10413209508406961. [DOI] [Google Scholar]
  36. Svebak, Sven , and Murgatroyd Stephen. 1985. “Metamotivational Dominance: A Multimethod Validation of Reversal Theory Constructs.” Journal of Personality and Social Psychology 48(1): 107–116. 10.1037/0022-3514.48.1.107. [DOI] [Google Scholar]
  37. Szabo, Attila . 2013. “Acute Psychological Benefits of Exercise: Reconsideration of the Placebo Effect.” Journal of Mental Health 22(5): 449–455. 10.3109/09638237.2012.734657. [DOI] [PubMed] [Google Scholar]
  38. Szabo, Attila. 2023. “Placebo Doping in Sport: Overview and Ethical Considerations.” The Journal of Sport and Exercise Science 1(1): 60–67. 10.36905/jses.2023.01.08. [DOI] [Google Scholar]
  39. Szabo, Attila , and Müller Anna. 2016. “Coaches’ Attitudes towards Placebo Interventions in Sport.” European Journal of Sport Science 16(3): 293–300. 10.1080/17461391.2015.1019572. [DOI] [PubMed] [Google Scholar]
  40. Weimer, Katja , Gulewitsch Marco D., Schlarb Angelika A., Schwille‐Kiuntke Juliane, Klosterhalfen Sibylle, and Enck Paul. 2013. “Placebo Effects in Children: a Review.” Pediatric Research 74(1): 96–102. 10.1038/pr.2013.66. [DOI] [PubMed] [Google Scholar]
  41. Wikimedia, Commons . 2023. File:09‐05‐2017 Gum Arabic (Acacia Nilotica) Seedpods (1). JPG: Retrieved from. https://commons.wikimedia.org/wiki/File:09‐05‐2017_Gum_arabic_%28Acacia_nilotica%29_seedpods_%281%29.JPG. [Google Scholar]
  42. World Medical Association . 2013. WMA Declaration of Helsinki—Ethical Principles for Medical Research Involving Human Subjects. 64th WMA General Assembly. Fortaleza, Brazil: Retrieved from. https://www.wma.net/policies‐post/wma‐declaration‐of‐helsinki‐ethical‐principles‐for‐medical‐research‐involving‐human‐subjects/. [Google Scholar]

Articles from European Journal of Sport Science are provided here courtesy of Wiley

RESOURCES