Spectator noise is a common factor at most major sporting events, and spectator noise has been shown to affect the performance of athletes and technical officials in terms of performance and decision-making (Myers & Balmer, 2012; Nevill et al., 2002; F. Sors et al., 2019; Starcke & Brand, 2012; Unkelbach & Memmert, 2010).

Badminton is no exception, and high spectator noise levels are common at badminton championships (Sjödin & Fahlström, 2018). Research has shown that spectator noise can affect referees' decision-making in football (Nevill et al., 2002; Sors et al., 2021). However, the reason for this effect can vary from person to person (Nevill et al., 2017) but also that that the decisions of referees with high anxiety might be more easily influenced by external factors like spectator noise (Sors et al., 2019).

High noise levels may in itself be perceived as unpleasant and thus cause stress and secondly, especially in a sports context, make it challenging to perform the work task and thus cause stress (Burns et al., 2016; Goyal et al., 2010; Fabrizio Sors et al., 2019). In a sports context, with spectator noise as the primary noise source, noise exposure is also more variable and subject to less control for technical officials such as umpires. Lower control of the noise source has been shown to be associated with lower cognitive performance (Belojevic et al., 2003; Paunović et al., 2009; Sobotova et al., 2010; Wright et al., 2016).

It is also likely that the vast media and spectator interest is putting additional pressure on the technical officials at the various venues. The time they spend working in the noise is also of great importance. Longer exposure time has been shown to increase fatigue (Ackerman et al., 2010).

During high-attention work at a high level of competition, cognitive resources can become depleted. It is argued that this affective experience of exhaustion can place additional cognitive demands on the individual as they attempt to maintain attention levels. If this continues during work, performance may decline to an unacceptable level (Beal et al., 2005). This is thought to harm our ability to maintain our attention on the work task, especially if we have had fluctuations in performance during a workday. It is suggested that this may harm our ability to maintain our attention on the work task, especially if we have previously had challenging tasks during a previous workday (Trougakos & Hideg, 2009). Working with a high cognitive load, pressure from spectators, and a high noise level that strains working memory (Baddeley, 2001) can lead to cognitive fatigue, so referees tend to make more mistakes over time (Nevill et al., 2002; Nevill et al., 2017).

The physiological explanation for the stress response due to noise is that a physical and mental imbalance caused by noise triggers a stress response to maintain homeostasis. The sympathetic nervous system becomes hyperactivated with chronic stress, leading to physical, psychological, and behavioral abnormalities. Studies have shown that noise exposure in the workplace can lead to hormonal changes through the release of cortisol (Ising & Kruppa, 2004; Ising & Prasher, 2000; MIKI et al., 1998). Measuring stress in an objective, non-intrusive way is difficult, especially in sports. As mentioned earlier, stress can be measured by taking cortisol samples, but this method is time-consuming, costly, and challenging to conduct during an activity. Questionnaires can also be used but do not measure the stress experienced during the activity. Heart rate variability (HRV) is another way to measure stress and has been used as an objective measure of stress in several studies (Jiménez Morgan & Molina Mora, 2017; Pagaduan et al., 2020).

It has been reported (Dimitriev & Saperova, 2015) that psychological stress increased predictability, regularity of RR (R waves) intervals, and reduced complexity. This reflects a shift towards more stable and periodic HR (heart rate) behavior under stress. The results of (Sim et al., 2015) showed the impact of different types of noise on HRV parameters. Reductions in HRV were observed during noise exposure: After adjusting for noise frequency, during low-frequency noise exposure, HF was reduced (a reduction of 32%, 34%, and 16%, respectively), and during high-frequency noise exposure was found to reduce LF (low frequency) by 21% compared to no noise exposure (Walker et al., 2016). The most commonly reported factor associated with variations in HRV variables was low parasympathetic activity, characterized by a decrease in the high-frequency band and an increase in the low-frequency band. Neuroimaging studies suggest that HRV may be associated with cortical regions (e.g., ventromedial prefrontal cortex) involved in stress appraisal; recent neurobiological evidence suggests that HRV is influenced by stress and supports its use for objective assessment of stress (Kim et al., 2018).

Considering the complexity of working as an umpire in badminton, which requires quick decision-making, it is interesting to investigate to what extent spectator noise can influence the umpire's experience of stress. In badminton, the umpires have two functions during matches: the main umpire and the service judge. The main umpire sits above the court and is responsible for all decisions on and around the court. The umpire ensures that the game is played according to the badminton rules. The main umpire makes calls regarding service errors and other errors or omissions of the players. The umpire keeps the score during the match and records all incidents of misconduct. The umpire's jurisdiction extends from entering the court before the match until leaving the court after the match. The second function is the service judge, whose task is to judge the technical execution of the player's serve and to assist the main umpire in making a decision in case of uncertainty. The service judge sits on the floor next to the net to fulfill these tasks. The two different roles are of particular interest due to their difference in cognitive load during matches. There is a lack of studies investigating the effects of noise on stress, especially in a sports context, using objective data on stress levels in an ecologically valid setting.

AIM

The study aims to use HRV data from umpires at a major badminton tournament to investigate whether there is an associationbetween audience noise and stress reactions and whether this possible association is also related to subjective noise sensitivity.

Secondly, the study also aims to investigate whether a possible association between spectator noise and stress differs depending on the officiating role (i.e., a main umpire and service umpire).

MEASURES

Heart rate variability measurement

Three kits of eMotion Faros 180 model for heart rate variability were used in the present study. The kits were mounted on the participating umpires in the morning before their first match every day of the tournament. Due to the limited number of kits available, all matches could not be measured for all umpires. The kits were evenly distributed among the participating umpires during the whole tournament. After the umpires last match of the day, the equipment was removed. If other participating umpires, not wearing a kit, had more matches the same day, the kit was mounted on that umpire to maximize the number of measurement hours.

Figure 1 Positioning of the ECG-electrodes and the recorder on the trunk

The ECG signals were recorded (1000 Hz) offline over a defined period of time (day, hours), events (breaks, umpire-times etc.) were logged, and after review transferred to KUBIOS-HRV-Software. This is a full-featured hear-rate-variability analysis software: it was used for accurate QRS and pulse wave detection, automatic artifact correction algorithms, automatic analysis sample generation, the calculation for all commonly used time-domain, frequency-domain and nonlinear HRV analysis parameters and for time-varying.

In Table 1 are the descriptive data regarding heart rate variability data (PNS = parasympathetic nervous system, SNS = Sympathetic nervous system, Stress = software calculation generating a stress value based on the relation between PNS and SNS) as well as the subjective noise sensitivity (NSS-11) and subjective stress (PSS-10) presented.

The correlations in Table 2 show that there is a statistically significant correlation between measured stress using heart rate variability and LaFMax (A weighted maximum sound level) and when working as an umpire r(1314) = .08, p = .006 and as a service umpire r(1058) = 0.27, p = .001. A similar result was also observed regarding dBALeq (A weighted equivalent sound level) and stress when working as an umpire r(1314) = .17, p = .001, and stress when working as a service umpire r(1058) = .029, p = .001. A statistically significant correlation was also observed for stress and Cpeak (C weighted impulse sound) when working as a service umpire r(1058) = .023, p = .001. However, no statistically significant correlation was observed for Cpeak when working as an umpire.

The difference in strengths of the correlations for umpires and service umpires was tested using Fischer's r to z transform the R values, presented in Table 2. By comparing the Z scores, the analyses revealed that there is a statistical difference, P < .05, for Laf_Max, Z_observed = -4.8, dBA_Leq, Z_observed = -3.1, and C_peak, Z_observed = -3.5 between the two groups. The results show that when working as a service umpire, there is a stronger association between noise levels and stress than when working as an umpire.

Correlations at an individual level for working as an umpire or a service umpire are presented in Table 3. As can be observed, some individuals, especially when working as a service umpire, show a stronger correlation between noise levels and stress. As can be observed for Umpire 1, the stress levels increased when exposed to higher LafMax noise levels as well as equivalent dBA noise levels when working as a service umpire. A similar result was observed for Umpire 5. Umpire 4 showed no statistically significant correlations between noise and stress levels.

Based on the individual results in Table 2, a Mann-Whitney U test was used to analyze whether umpires have an association between noise exposure (dBALeq) and stress (HRV) during matches working as a service umpire (Umpire 1, 2, 5, 7, 9) differed regarding their rated noise sensitivity and general stress different compared to umpires (Umpire 4, 6) not having this association. Noise sensitivity data is missing for umpire 3 and 8. Umpires with an association between noise and stress during matches reported higher subjective noise sensitive (Mdn = 36.0, n = 5) compared to the non-association group (Md = 27.5, n = 2), U = 3.5, z= -0.61, p >.05, however; the difference was not statistically significant.

When comparing the same groups regarding general experienced stress, the results showed that umpires with an association also reported their general stress as higher (Mdn = 15.0, n = 5) compared to the non-association group (Mdn=11.5, n = 2), U = 1.5, z= -1.37, p >.05, however; the difference was not statistically significant.