Participants comprised of volunteers from Loughborough University. Thirteen healthy males aged 24 ± 4 years (mean ± SD) (weight: 73 ± 11 kg; height: 178 ± 5 cm), that satisfied the minimum physical activity requirement (30-min moderate intensity, physical activity on at least 3 days each week for at least 3 months) with no history of neurological or auditory disability were recruited. Participants were paid for their participation pro rata. The study was approved by the Loughborough University Ethics Committee (R17-P190) and was conducted in compliance with the Declaration of Helsinki except registration in a database (2013).

All experiments were conducted in the Environmental Chamber (T.I.S.S. Peak Performance, Series 2009 Climate Chambers) at the Environmental Ergonomics Research Centre, Loughborough University, United Kingdom. Participants were requested to come in at the same time of day for each trial in a similar physical state as the previous trial. Participants were also requested not to consume any food or caffeine for up to 2-h prior to participation in the study. Trial order was counterbalanced with a minimum of 48-h between trials. In total, the current study consisted of five trials, in which four main experimental trials followed an initial familiarization session.

The present study incorporated the use of self-selected motivational music. Prior to attending the familiarization session, participants were asked to self-select their top 10 motivational songs. They were asked to choose songs that would motivate them during exercise where “motivate” meant “make you want to exercise harder and/or longer” (Karageorghis et al., 2006). In the familiarization session participants rated their chosen songs using the Brunel Music Rating Inventory-2 (BMRI-2). This inventory rates music based on the rhythm, style, melody, tempo, sound, and beat of the music on an agreement scale from 1 (strongly disagree) to 7 (strongly agree). The top seven rated songs were chosen as the most motivational for each individual participant and subsequently used in the main experimental trials.

Following the music selection protocol, participants were accustomed to the experimental procedures and a full rehearsal of the physical performance test was completed. The physical performance test included a time trial, as well as a series of brief and sustained isometric neuromuscular tests, including supramaximal nerve stimulation. For full details of the physical performance protocol refer to section: “2.4 Physical Performance Protocol.” This concluded the familiarization session.

The present study consisted of four [FiO₂ (fraction of inspired oxygen) × 2 and music; ×2] main experimental trials. The trials included: (1) normoxia (sea level, 0.209 FiO₂) and no music; (2) normoxia (0.209 FiO₂) with music; (3) normobaric hypoxia (∼3800 m, 0.13 FiO₂) and no music; and (4) normobaric hypoxia (0.13 FiO₂) with music. Normobaric hypoxia was manipulated using the inbuilt hypoxic air generator in the environmental chamber (T.I.S.S. Peak Performance, Series 2009 Climate Chambers). A Servomex gas analyzer (570A, Sussex, United Kingdom) was used to monitor the oxygen level inside the chamber, as well as the inbuilt analyzer on the chamber.

All experimental trials were completed at 21°C with 50% relative humidity. In each of the main experimental trials, participants completed a protocol (Figure 1) that consisted of: (1) 15-min time trial on an arm-bike, followed by (2) a 60-s isometric maximal voluntary contraction (MVC) of the biceps brachii with supramaximal nerve stimulation. Participants were blinded to the environmental condition.

All exercise testing was conducted on a custom-made arm-bike rig, built by the Environmental Ergonomics Research Centre. The device was developed to measure arm bike performance and facilitate a fast transition to measure neuromuscular function of the biceps brachii. Therefore, the device consisted of a seat and chest harness, a hand bike ergometer (Lode Angio B.V., Groningen, The Netherlands) located in front of the participant, and on the side, a right arm support which included an adjustable force transducer and wrist strap (O’Keeffe et al., 2021).

In each experimental condition, upon entering the chamber, participants practiced a series of 3-s MVC’s with supramaximal nerve stimulation. Participants then conducted a 15-min time trial on the arm bike where they were instructed to perform as much work as possible. Participants did not receive any external reinforcement for the duration of the physical performance protocol and were only informed when they had 5-min remaining in the time trial. This mode of exercise performance was used to represent self-regulated upper body work which is characteristic of high-altitude pursuits. Upper body work is particularly dominant in climbing sections, with scrambling (ascending steep terrain using hands), technical climbing using ropes, hand holds, ladders, ice axes and carrying, pushing, and pulling loads all prominent features. In addition, a time trial performance component is representative of the self-regulated work and further, the arm bike ergometer was used, as the speed of transition from the ergometer to the custom-built rig for performance of the MVC was important to ensure accuracy of the neuromuscular assessment.

The resistance factor on the arm bike ergometer was determined based on the linear relationship with a traditional monarch brake weight and is defined as; resistance factor alpha = 0.0161 × brake weight + 0.0008. The brake weight corresponds to 1% of a participant’s familiarization body mass (73 ± 11 kg). This particular brake weight was chosen based on pre-experimental piloting and previous research conducted in the laboratory using this method (O’Keeffe et al., 2021). By incorporating this method, performance in the time trial (i.e., power output) is therefore dependent on cadence.

The self-selected motivational music was compiled into playlists for each individual participant and played through two speakers (Mission, United Kingdom) connected to an integrated stereo amplifier (Sony, TA-FE230), placed directly in front of the arm-bike rig (1.2-m from the participants head). Music was played at a standardized noise level of 80-dBA (at participant’s ears) monitored using a sound level meter (Model: LxT 831, Larson Davis Inc., United States) as per previous research indicating that 80-dBA enhances exercise performance (Edworthy and Waring, 2006; Karageorghis et al., 2018). Selected music ranged from slow-paced church music to faster-paced hip hop and heavy metal. The music was played from the start of the 15-min time trial and stopped at the end of the MVC.

Following the 15-min time trial, participants performed a 60-s MVC with simultaneous nerve stimulation every 10-s. Supramaximal nerve stimulation (using the doublet interpolation method) was used to quantify central and peripheral fatigue (Merton, 1954; Herbert and Gandevia, 1999). Pre-and post-exercise, during the MVC of the right bicep brachii, electrical impulses (300V, doublet square-wave, with a 10-ms stimulus interval) were delivered using a nerve stimulator (DS7AH, Digimeter Ltd., United Kingdom). The stimulator anode was placed over the medial border of the right scapula, and the cathode was placed on the distal tendon of the right bicep. The optimal position for stimulation of the biceps brachii was confirmed based on the force output of the biceps in response to submaximal current stimulation (∼50 mA) and then progressively increased by 10 mA until a plateau in the mechanical response of the muscle was observed (91 ± 25 mA).

After securing the right arm in the arm support, two stimulations were first given at rest. Following this, participants performed a 60-s sustained MVC in which superimposed doublet twitches were evoked (Qtw,_sup) every 10-s. The sustained MVC was immediately followed by two further resting potentiated doublet twitches (Qtw,_pot) 1-s after full muscle relaxation. Voluntary muscle activation of the bicep (VA%) was calculated using the equation: VA% = (1 − Qtw,_sup/Qtw,_pot) × 100 (Folland and Williams, 2007) (where Qtw,_sup is the average of doublet twitches evoked over the 60-s MVC).

Two-way repeated measures ANOVA was used for the analysis of the main effects and differential impact of music and hypoxia on physical performance (2 × 2: FiO₂ × music). Within trials, pre-test and post-test were compared using paired sample t-tests with Bonferroni correction for multiple comparisons. To calculate the power required to observe a main effect of hypoxia (2 levels) and music (2 levels), a power analysis conducted in G*Power (Faul et al., 2007) incorporating a large effect size (f = 0.4), power of 0.8 with an α err prob of 0.05 yielded a sample size of 12. The power required for a one-tailed (pre to post) t-test using a large effect size of d = 0.8, a power of 0.8, and an α err prob of 0.05 outputted a required sample size of 12. A 95% confidence level was used to test significance (p < 0.05). When Mauchly’s Test of Sphericity was significant (p < 0.05), the Greenhouse–Geisser adjustment was used. Effect sizes, taken from SPSS, are reported as partial eta squared (ηp2) for main effects and interactions. Data is presented as mean ± SD.

Average power output (W) was reduced with a main effect of hypoxia (p = 0.02, ηp2= 0.63) and significantly increased with a main effect of music (p = 0.001, ηp2= 0.38) (Figure 2A). When combined the interaction was additive (p = 0.87, ηp2= 0.002). This indicates that the effect of one factor (e.g., FiO₂) was not altered by a change in another factor (e.g., music). Maximum power output was unaffected by hypoxia (p = 0.08, ηp2= 0.23) but was significantly increased with a main effect of music (p = 0.01, ηp2= 0.42) (Figure 2B). The pacing strategies for each trial are presented in Figure 3. A main effect of music, where music increased average power output was observed from minute 1.5 until minute 13.5, p ≤ 0.04. Further, a main effect of hypoxia, where hypoxia decreased average power output was observed from minute 6 until minute 15, p ≤ 0.04 (Figure 3).

One participant was removed from the MVC analysis due to technical difficulties. During the post-exercise MVC, average MVC force (N) was reduced in hypoxia (p = 0.03, ηp2= 0.35). No effect of music was observed on average MVC force, p = 0.53 (ηp2=0.03) (Figure 2C). Similarly, peak MVC force (N) was reduced in hypoxia (p = 0.03, ηp2 = 0.35) and no effect of music was observed (p = 0.53, ηp2 = 0.03) (Figure 2D). No effect of hypoxia was observed on VA% of the bicep brachii, p = 0.47 (ηp2=0.05). However, a main effect of music, where music increased VA% in both normoxia (55.32 ± 27.82%) and hypoxia (47.87 ± 28.7%), p = 0.02 (ηp2=0.4), was observed (Figure 4). No interaction was observed indicating an additive effect, p = 0.96. Post-exercise resting potentiated doublet twitch force (N) was unaffected by hypoxia (p = 0.95) or music independently (p = 0.99), with no interaction observed (p = 0.61) (Figure 4).

Two participants were removed from the analysis of NIRS due to technical errors. A main effect of hypoxia was observed on muscle deoxygenation of the bicep where hypoxia increased muscle deoxygenation (HHb) (p = 0.05, ηp2=0.34). Music had no effect on muscle deoxygenation, p = 0.40 (ηp2=0.07) (Table 1). Furthermore, no effect of hypoxia (p = 0.7, ηp2=0.02) or music (p = 0.34, ηp2=0.09) was observed on oxygenation of the muscle (O₂Hb) or systemic (VO₂) oxygen consumption (p ≥ 0.12) (Table 1). No effect of hypoxia (p = 0.32), music (p = 0.29), or interaction (p = 0.51) was observed on heart rate.

Subjective ratings of discomfort and perceived effort are presented as an average over the 15-min time trial. Breathing (5.6 ± 1.5, ηp2=0.14) and arm discomfort (6.5 ± 1.7, ηp2=0.25) were increased in the presence of hypoxia, p ≤ 0.002, but music had no effect on breathing discomfort (ηp2=0.02) or arm discomfort (ηp2=0.02), p ≥ 0.24. However, a significant interaction where music decreased subjective scores of breathing (5.2 ± 1.7, p < 0.001, ηp2=0.27) (Figure 5A) and arm discomfort (6.2 ± 1.81, ηp2=0.21) in hypoxia was observed (Figure 5B). Independently, no effect of hypoxia (p = 0.41, ηp2=0.01) or music (p = 0.36, ηp2=0.01) was observed on subjective scores of mental effort. However, when combined, a significant interaction where music reduced ratings of mental effort was reported (47.84 ± 17.25, p = 0.001, ηp2=0.16) (Figure 5C). A main effect of hypoxia was observed on RPE (15 ± 2, p < 0.001, ηp2=0.21), where ratings were increased (Figure 5D). However, no effect of music was seen on RPE (p = 0.42, ηp2=0.01), and no interaction was observed (p = 0.06, ηp2 = 0.05).

Motivation questionnaires were analyzed in terms of task motivation, success motivation, and overall motivation, and were measured pre-exercise and post-exercise. There were no significant differences in pre-task motivation scores (p = 0.32), pre-success motivation scores (p = 0.56), or pre-overall motivation scores (p = 0.17).

Motivation scores were analyzed in relation to percentage change from pre-test to post-test (Table 2). No effect of music (p = 0.30, ηp2=0.08) or hypoxia (p = 0.20, ηp2=0.13) was observed on success motivation, with no interaction presented (p = 0.68, ηp2=0.02). Further, there was no effect of music (p = 0.19, ηp2=0.13), hypoxia (p = 0.07, ηp2=0.25), or interaction (p = 0.79, ηp2=0.01) on task motivation. Likewise, overall motivation was unaffected by music (p = 0.75, ηp2=0.01) and hypoxia (p = 0.29, ηp2=0.09), with no interaction (p = 0.67, ηp2 = 0.02).

Regarding mood, there were no significant differences in pre-test scores across all moods, p ≥ 0.23. Likewise, across trials, there were no main effects of hypoxia (p ≥ 0.11) or music (p ≥ 0.15) on any post-test subjective mood scores (Table 2).

In general, the use of the hand bike ergometer results in lower VO₂ (ml⋅kg^–1⋅min^–1) scores observed largely due to the specificity of the muscle groups that are involved and the lower amount of active muscle tissue as compared to other modes of exercise (e.g., rowing and cycling). The primary working muscles used in this exercise mode are the biceps, triceps, and deltoid (Mitropoulos et al., 2017) which are naturally smaller and less conditioned for endurance exercise. Nevertheless, in the present study, music increased power output over a 15-min time trial in both normoxia and hypoxia. In normoxia, the ergogenic benefits of music on exercise performance have been well established (Karageorghis and Priest, 2012a,b; Terry et al., 2020) and support the finding of the current study. In hypoxia, however, no previous research has been conducted on whether similar ergogenic effects of music on performance can be seen. In general, in the context of extreme environments, research investigating the impact of music on exercise performance is limited. Of this research, English et al. (2019) investigated the impact of music on a 15-min time trial on a bike ergometer in heat stress (36°C). They found that with self-selected motivational music, total work output in the heat was increased by ∼10%. Results from the current study corroborate these findings as, in the presence of hypoxia, exercise performance was increased by 6.4% and similarly in normoxia, exercise performance was increased by 6.3%. Further research is warranted to explore the impact of music as an intervention in other extreme environmental conditions (i.e., cold and rain), and to investigate the differential impact of music in greater levels of hypoxia (i.e., >3800 m), where the decrements in physical, cognitive, and psychological constructs may be exacerbated (Bahrke and Shukitt-Hale, 1993; Goodall et al., 2012; Komiyama et al., 2017).

A unique element of the present study involved participants completing a 60-s isometric voluntary contraction with simultaneous nerve stimulation to quantify central and peripheral fatigue using the doublet interpolation method. In the presence of hypoxia, maximal voluntary force was reduced when compared to the normoxic control. However, when combined, music increased the VA% of the bicep brachii. This is the first study to demonstrate that self-selected motivational music increases VA%, i.e., neural drive to the muscle, mitigating the onset of central fatigue in both normoxia and hypoxia. Limited research investigating the impact of music on muscular endurance exercise exists. One such study investigating the effects of motivational music on a muscular endurance task (isometric weight-holding task with a 1.1 kg dumbbell) observed that participants were able to sustain the task for 11% longer when motivational music was played throughout the task (Crust and Clough, 2006). Another study investigating the effects of music tempo and intensity on grip strength found that fast tempo (126 bpm) and high intensity (80 dBA) music increased grip strength coupled with greater subjective scores for positive affect and arousal (Karageorghis et al., 2018). Whilst these studies indicate the potential for music to improve muscular endurance performance, they did not measure neuromuscular activity and hence, further research is warranted to support the current findings that self-selected motivational music increases VA% in normoxia and hypoxia.

Significant advances have been made toward understanding the mechanisms underpinning the effects of music on exercise performance (Terry et al., 2020). Rejeski’s (1985) model of parallel information processing has been widely used and accepted as an explanation to how music impacts exercise performance. In this model, afferent feedback to the central nervous system is impeded by music whereby, music has the capacity to flood the physiological feedback signals which are associated with physical exertion and are of limited processing capacity (Rejeski, 1985; Karageorghis and Priest, 2012a). It is thought that music, through the facilitation of pleasurable stimuli to sensory pathways in the brain, may alter the perception and interpretation of fatigue and hence distract or block sensations of pain and fatigue from focal awareness (Karageorghis and Priest, 2012a). This mechanism can explain the impact of music on low to moderate intensity exercise as individuals are better able to redirect attention to task-irrelevant cues promoting positive affect. However, as the intensity of exercise increases, attentional focus in forced back toward the internal monitoring of task-related cues, i.e., fatigue (Bigliassi et al., 2016). In the aforementioned study, English et al. (2019) suggested that the facilitative nature of music in their findings may be due to the interference of music with afferent feedback of fatigue and pain caused by the exercise and heat strain. Interestingly, they found that music facilitated performance in this capacity up until the last two time blocks where there was a steady decline in self-selected exercise intensity. This may have been due to the exercise and fatigue overriding the motivational music, and it also may be due to the increased thermal perception associated with the relative heat strain which further could be considered as a cognitive load indicating the initiation of a thermoregulatory behavioral response (i.e., reduced exercise intensity) (Gaoua et al., 2012). In the current study, however, the music trials elicited a greater average power output throughout the 15-min TT, which may be due to music overriding the pain and fatigue of the exercise, but also the hypoxic strain.

In this respect, previous research reporting the impact of music on RPE have resulted in inconsistent findings (Elliott et al., 2005; Karageorghis and Priest, 2012a). It has been suggested that music facilitates greater exertion whilst distracting participants from the impact of this increased work output (Karageorghis and Priest, 2012a) hence maintaining RPE. The results from the current study support this suggestion, that RPE remained largely unchanged due to the mechanism of dissociation, a coping mechanism, which facilitates an individual in blocking out the increased discomfort in hypoxia and normoxia. Similarly, previous research has demonstrated that music enhances mood and motivation during exercise (Tenenbaum et al., 2004; Elliott et al., 2005; Karageorghis and Priest, 2012a). In contrast, the current study found music to have no effect on post-exercise mood or motivation. A possible explanation, however, could be that much of this research has recorded mood and motivation change throughout exercise, whereas the current study monitored change from pre- to post-exercise. Considering music, independently, had no impact on ratings of discomfort, mental effort, RPE, mood, and motivation, these results provide further evidence for the theory of dissociation whereby music maintained subjective scores despite a higher power output (i.e., increased pacing).

The current research was conducted in two different environmental conditions including normoxia (0.209 FiO₂) and normobaric hypoxia (0.13 FiO₂). A potential limitation of this research and a future direction to consider is that a higher level of hypoxia was not included in the experimental trials, e.g., >0.13 FiO₂, which would provide a greater insight into the differential impact of music on physical performance in hypoxia. It has previously been reported that at moderate hypoxia, peripheral fatigue is the primary contributor of the decrease observed in the performance of small muscle mass exercise (Millet et al., 2012; Fan and Kayser, 2016; O’Keeffe et al., 2021). Conversely, in more severe hypoxia, it has been observed that central fatigue is the primary factor impacting performance via alterations in cerebral oxygenation (Millet et al., 2012). Incorporating a measure of cerebral oxygenation could therefore facilitate a greater interpretation of the impact of music on cerebral dynamics in hypoxia. Further, another consideration for future research includes investigating the impact of music on physical performance in hypobaric hypoxic conditions, which may yield different results.

This study used an arm bike ergometer to represent self-regulated upper body work in normoxia and hypoxia. This was a relatively unfamiliar task to participants, however, utilizing the arm-bike effectively is not difficult. To use the arm-bike, participants were required to push with one arm and pull with the other. All participants attended a familiarization session which included a practice of the entire physical performance protocol prior to undertaking the main experimental trials.

A limitation of research investigating the impact of self-selected music on performance outcomes, is that it is difficult to completely blind participants to the objectives of the study. In the current study, participants were asked to select songs that would motivate them to exercise harder and/or longer as per the guidelines of the BMRI-2 (Karageorghis et al., 2006). Hence, the issue lies in the potential of participants to alter their behavior in the presence of music. Despite this, due to music having no impact on subjective scores (i.e., RPE, discomfort, mental effort, motivation, and mood) we can conclude that participants were unaware of their greater exertion in relation to their increased power output in the time trial and hence, did not alter their behavior as a result of not being blinded to the music component.

With regards to the subjective ratings of perceived exertion, participants were asked to rate their overall exertion on a whole-body scale, which differs to a localized peripheral rating of perceived exertion. As proposed by Hutchinson et al. (2019), individuals unfamiliar or untrained in upper body exercise such as the arm crank, should provide differentiated scores of RPE, i.e., peripheral RPE (arms) and overall RPE (whole-body) to gain a more accurate indication of the exertion of the participant. In the current study, participants untrained in arm-crank exercise were used and therefore whole body RPE may not have been a true representation of their exertion.

Much debate surrounds the use of the doublet interpolation method as a reliable and valid measure for the quantification of muscle activation. Despite its limitations however, this method is widely used to quantify central and peripheral fatigue in maximal conditions (Shield and Zhou, 2004; Folland and Williams, 2007; Lloyd et al., 2016) and is especially useful when comparing muscle activation within-subjects, in a repeated measures design such as the one incorporated in the current study. This study utilized a 300V, doublet square-wave (with a 10-ms stimulus interval) which have been shown to be a better estimate of muscle activation than a single stimulus (Folland and Williams, 2007). Further, participants were harnessed and strapped into the custom-built neuromuscular rig specific to each individual’s height and limb length to ensure optimal and consistent angular positioning. Each participant’s individual stimulus was determined through a ramped protocol in the familiarization session (i.e., the stimulus strength when a plateau in the mechanical response of the muscle was observed), and the electrode placements subsequently marked for consistent stimulation. Baseline doublets were performed prior to the beginning of the physical performance protocol to ensure the correct placement of the electrodes. However, they were not recorded and used for comparative analysis and therefore a further limitation of this study is that the resting doublets were performed after the time trial, prior to performing the MVC, and hence, may already be in a fatigued state.

Lastly, a major limitation of this study is that it included males only. However, research has demonstrated that females may be more susceptible to different psychological constructs at altitude, where sex has been shown to be an independent predictor of significantly heightened anxiety at altitude (Boos et al., 2018). As this study was focused on investigating music as a mechanism to alleviate performance decrements and further, observe mechanisms facilitating the variation in responses to hypoxic environments, focusing on one gender allowed greater clarity in the first instance. However, research is warranted to investigate the psychological constructs which may help to explain some of the variation observed in physical performance, and further alleviate the performance decrements being observed in altitude amongst females.