Introduction
Music-related interventions have been widely used in sports and exercise (3,9,16) as a way to improve performance, decrease perceptions, or modulate physiological parameters (26,37). There are numerous external factors that might alter and contribute to the right recommendation and method for using music to exercise (18,19). As an example of this, the time of stimulus application, the level of synchronization, the music tempo/expression, and the type of song may act differently inside the brain, causing a variety of outcomes. All of these components should be included in a conceptual framework for each specific situation. However, in practical terms, music is applied in exercise using self-selected method on account of individuality and psychological differences of humans (4). Furthermore, this method is easier to consider because exercisers can choose appropriate songs by themselves and exercise according to their preferences. Based on this assumption, testing self-selected music represents more practical applications compared with other methods.
Running is a common physical exercise worldwide because of its own features (low cost and availability for practice) and high aerobic benefits (21). In addition, music has constantly been applied for this kind of exercise (8,17), and this task represents an important method for studying the effects of music on psychophysiological (36,42) and cerebral parameters (29). The use of music in running is based on psychological theories that support the use of music in exercise. Previous researches have demonstrated that music can aid running by acting in parallel to exercise, using the dissociation system (32,33) or increasing emotional responses (↑ cerebral catecholamines) (10,11).
In 1998, Szmedra and Bacharach (36) identified interesting responses from the use of music on psychophysiological variables. The protocol consisted of 15 minutes of running at 70%
;)
. Among the study's findings, they identified lower autonomic, physiological, and psychophysical responses in the music condition (classical) compared with the control. Thereafter, several studies initiated applying other physiological assessments during music-related interventions, which demonstrate that auditory stimulus is likely to affect the “body as a whole” even during exercise conditions.
It is also important to highlight that music may be applied through a mechanism of sedation; for instance, at the end of severe physical exercise, the recovery process is affected by numerous variables, and accelerating this component allows prevention. Eliakim et al. (9) have shown that psychological strategies such as music after exercise can ameliorate fatigue-related symptoms (perceived exertion; p ≤ 0.05) and physiological responses (lactate; p ≤ 0.05). They played motivational tracks in an attempt to increase vigor, the number of steps, and subsequently blood lactate removal. Laboratory studies have constantly increased our knowledge regarding music-related interventions. Nevertheless, real training or competition situations using music are still scarce in the literature, and the outcomes of previous research might be dramatically changed during actual conditions of exercise.
Based on the previous statements, what would be the real effects of music on performance, psychophysical, physiological, and recovery variables with 5 km of running? Would amateur runners be capable of choosing motivational songs? Would music be capable of changing emotional responses (↑ autonomous system activity) and cerebral activation? In this study, all of these questions form the main goal.
Our hypotheses are supported by the past research that showed the effect of music based on task intensity and physical fitness level. By testing music in 5-km running, we have created a real outdoor model representing a training situation. This design represents an endeavor to understand and recommend music in exercise. Hence, we expect a slight improvement in performance parameters when the sensation of fatigue is not already present, as well as the maintenance of psychological conditions because of its high intensity. Finally, modulation of the autonomous system might occur, according to Conrad et al. (7) who suggest that sedative songs can act in conscious and unconscious cortical areas, increasing pituitary activity and growth hormones, followed by less stress modulation (↓ interleukin-6, ↓ epinephrine, ↓ dehydropiandrosterone) causing lower blood pressure and a decreased heart rate. These outcomes are presumed to happen because of an indirect effect on the sympathetic nervous system, decreasing inflammatory reactions.
Methods
Experimental Approach to the Problem
This study was divided into 3 stages that were performed in the course of 30 weeks. In the first stage, all participants were interviewed separately before the experiment. At this time, they gave their anthropometric measures (weight and height), personal information (age, time of continuous training, number of running competitions, and training volume), and answered the Eysenck personality questionnaire (EPQ) (1), which gives possible stratifications according to personality, checking whether music could act differently in accordance with personal features.
All subsequent procedures were explained, and they were familiarized with the rate of perceived exertion (RPE 15 points and CR10) (5) and the Brunel Mood Scale (BRUMS) (22). They were requested to select 30 motivational songs (10 slow-speed tracks, 10 medium-speed tracks, and 10 fast-speed tracks), and the only provided information was to select songs capable of increasing their vigor and motivation to accomplish a severe aerobic physical exercise (4).
When the number of tracks did not achieve the required number, they were asked to choose other songs to complete the playlist. The song stratification was performed initially by 2 applications (BPM Counter; AbyssMedia Freeware Software, Ulyanovsk, Russia, and Virtual DJ; Digital DJing market, Sevenoaks, Wealden Place, UK) followed by the examination of an expert musician.
In the second stage, all participants were instructed to be in the laboratory after the interview to perform a neuroimaging test involving listening to a variety of songs. This technique was conducted to demonstrate how self-selected songs could act in emotional areas of the brain, ensuring by physiological assessments the effectiveness of motivational music in inducing emotional consequences (24), and this topic is better described below.
Finally, all participants performed 5 physical tests individually (5-km run as fast as they could) in the first line of an official track (12 laps × 400 m + 200 final meters). The time between each test ranged from 72 (3 days) to 168 hours (7 days). All tests were performed at the same time of the day (±0.25 hours) to avoid circadian variation. In this study, 5 experimental conditions were tested (PM: motivational songs, ranging from 110 to 150 b·min−1, applied before 5 km of running; SM: slow motivational songs, ranging from 80 to 100 b·min−1, applied during 5 km of running; FM: fast motivational songs, ranging from 140 to 160 b·min−1, applied during 5 km of running; CS: calm songs condition, applied after 5 km of running; CO: control condition, without intervention). It is also important to note that CO was considered as a baseline condition that was used for further comparisons. Moreover, we created a silent environment for CO because talking could act positively or negatively for some participants, altering psychophysiological variables randomly. The chosen music tempo average was “129.9 ± 7.33 b·min−1 for PM,” “89.1 ± 6.30 b·min−1 for SM,” “146.4 ± 5.08 b·min−1 for FM,” and “103.6 ± 7.76 b·min−1 for CS.” The participants only knew which experimental protocol they would do upon arrival, and experimental conditions were randomized using the following site: www.randomization.com. Participants on average trained twice a week with music.
Subjects
We selected 15 amateur runners (24.87 ± 2.47 years; 78.87 ± 10.57 kg; 178 ± 07 cm; 24.92 ± 2.79 kg·m−2; 4.85 ± 1.85 years of training; 7 ± 3.49 weekly training hours; 5.67 ± 2.85 participated competitions). We calculated the number of subjects using performance time as the main variable from Terry et al. (38), which presents a certain level of similarity, assuming a significance value of 0.05, statistical power of 0.80, ratio of 1, the mean of differences between 2 experimental conditions, and the SD from the experimental protocol (GPower 3.1; Heinrich Heine University of Düsseldorf, Düsseldorf, Germany). Participants were instructed to refrain from vigorous activities, ingestion of caffeine, medication, alcohol-containing substances, and parallel ergogenic aids for at least 24 hours before the tests. Furthermore, they were instructed to maintain their nutritional habits and only to perform the physical tests in normal psychological conditions (avoiding personal interferences). These requirements represent an interesting way to control the real effects of music on exercise without being affected by external interferences. This study was approved by the local Institutional Research Ethics Committee, and all participants provided written informed consent approved by the University Institutional Review Board.
Procedures
Functional Near-Infrared Spectroscopy Assessment
A functional near-infrared spectroscopy (fNIRS) device (16 channel forehead sensor, 10 photodetectors, 4 photoemitters, 2.5-mm interoptode distance; BIOPAC Systems, Inc.; Goleta, CA, USA) was used to obtain the cerebral activation of 4 representative tracks (1 random track from each experimental condition). A technician applied the fNIRS sensors in line with positions FP1-FP2 on the International 10-20 system, designed for recording data from the prefrontal cortex (PFC) area. The device was positioned on the forehead 0.5 cm above the eyebrow, and all experimental procedures were performed in a dark room, avoiding external light interference.
Functional near-infrared spectroscopy is a neuroimaging modality that measures hemodynamic changes in the brain. Relative absorption and backscatter of near-infrared light by oxyHb and deoxyHb reflect changes in neural activity by neurovascular coupling. Furthermore, by lending itself to small, noninvasive, inexpensive, and portable lightweight devices, fNIRS technology is increasingly applied in functional brain imaging studies. The instrument was developed according to Chance and Leigh based on the modified Beer-Lambert Law (24). Each light source contained 2 light-emitting diodes with wavelengths of 730 and 850 nm, which represent the reflection of HHb (deoxygenated hemoglobin) and O2Hb (oxygenated hemoglobin) chromophores, respectively. The HHB and O2HB variation is considered a direct response to brain function through blood flow direction, and these events often occur when psychophysiological events are induced, provoking cerebral alterations (activity) and blood deviation (24).
Initially, participants remained seated and answered the Self-Assessment Manikin (SAM), including analysis of arousal and valence (6), ranging from +1 (minimal arousal/valence) to +9 (maximal arousal/valence). Afterward, a heart rate monitor was attached (Polar RS800; Polar Electro, Kempele, Finland) for heart rate variability (HRV) analysis. This topic is better described below. After that, they remained in silence, avoiding drowsiness until their values (O2Hb and HHb) showed a linear scattering, and the baseline was calculated for a period of 10 seconds. The procedure to consider baseline was white noise as background for 30 seconds (20 seconds before baseline calculation) (25). They refrained from caffeine and alcohol for at least 24 hours before the test, and the PFC was selected because of its cortical activation during the performance of emotional songs (28).
When this procedure was finished, they heard 4 songs (1 random track from PM plus 5 minutes of silence, 1 random track from SM, 1 random track from FM, and 1 random track from CS). The order of the tracks was randomized, and white noise was inserted between each one for a short period of time (30 seconds of clearance). The period of silence applied after PM represents an attempt to examine the residual effects of the music (25). It is important to note that a similar situation would happen during exercise (music applied before 5 km of running), and residual effects would be the only underlying mechanisms to provide exercise benefits. All tracks were played using an Ipod (Apple, Inc., USA) device and an Onbongo headset (Model ONB-M80; São Paulo, Brazil), keeping the volume at 75 (±5) dB measured directly on the ear with a decibel meter (DL-4020; Icel, Manaus, Brazil). Figure 1 represents an illustration of fNIRS application.
Figure 1: Illustrative representation of the experimental procedures regarding fNIRS application. fNIRS = functional near-infrared spectroscopy; HRV = heart rate variability analysis.
Experimental Protocols
This topic describes the outdoor experiment, which is detailed in Figure 2. The protocol was divided into 6 periods (arrival, resting, warm-up, 5-km running, recovery, and reevaluation), and all events are shown in Figures 2 and 3.
Figure 2: Illustration of the experimental protocol, including all physical tests of the study. *Water drink, restroom use, and cereal bar; **30 minutes after recovery. BRUMS = Brunel Mood Scale; Equi = equipment placement or equipment withdrawal; HRV = heart rate variability; SS = self-selected rhythm; CR10 = category ratio of effort; PM = previous motivational songs; SM = slow motivational songs; FM = fast motivational songs; CS = posterior calm songs.
Figure 3: Original picture of participant running with the headset.
Arrival
The participants remained seated for a short period of time (10 minutes), as they answered the BRUMS (pre). This tool (BRUMS) represents a questionnaire to assess mood states, consisting of 24 adjectives (0–16 Likert Scale) stratified among 6 mood domains: anger, confusion, depression, fatigue, tension, and vigor (22). Additionally, all equipment (heart rate monitor and pedometer—DigiWalker SW200; Yamax, San Antonio, TX, USA) were placed at this time, and they could spread sunblock (Sundown; Johnson & Johnson, New Brunswick, NJ, USA), repellent (Repelex, São Paulo, Brazil), drink water, and use the restroom if they judged necessary.
Resting
During the next 10 minutes, the participants remained lying and HRV analysis (described below) was performed, with a complete period of silence (when randomly selected songs from PM were inserted at this time).
Warm-up
The warm-up period consisted of 2 minutes of walking in a self-selected rhythm, plus 6 minutes of jogging at a heart rate of 130 b·min−1 (allowing ±5 b·min−1 variations), plus 2 minutes of running at a heart rate of 150 b·min−1 (allowing ±5 b·min−1 variations) (adapted from Skof and Strojnik) (34) in a synthetic space to run in circles (radius = ±10 m) in the middle of the track.
Five-Kilometer Run
The participants had to run 5 km as fast as they could, and for every lap (400 m) + 200 final meters they had to check their HR and RPE (15-point scale), answering the following question: “how hard is the task at this moment?” (5); during all tests, the values very, very light (number 7), and very, very hard (number 19) were used as anchoring points (27). They were also given instructions regarding the importance of the veracity of the reported RPE, highlighting the fact that answering a lower or higher value of perception would not influence their performance. In addition, the time of each lap + 200 final meters was recorded with a digital chronometer (Ipod; Apple, Inc.) by the same evaluator. For music conditions (PM, SM, FM, and CS), all tracks were played using the same previous device (another Ipod; Apple, Inc.) placed in a clamp (HammerHead; São José, Santa Catarina, Brazil) with a headset (Model ONB-M80; Onbongo), keeping the volume at 75 (±5) dB measured directly on the ear with a decibel meter (Icel DL-4020, Brazil).
Recovery
After 5 km of running, the subjects had to lie down immediately and stay there for a period of 10 minutes, collecting HRV information in a complete period of silence. For music condition (CS), all tracks (Enya, May It Be, New Age Style—110 b·min−1; Bach, Air, On the G String, Classic Style—106 b·min−1; Hilary Stagg, Pleasant Dreams, New Age Style—95 b·min−1) were chosen following style recommendations (41), and the track order was always the same.
Reevaluation
Finally, the participants remained seated and answered the BRUMS (post). All equipment was withdrawn, and they could drink water, use the restroom, and eat a cereal bar (Nestlé—banana taste; Nestlé S.A., Vevey, Canton of Vaud, Switzerland). At the end of 30 minutes, they answered the CR10 (10 points) concerning their general perception of effort in the whole period of running.
Heart Rate Variability: Running Test
The HRV was collected during the resting and recovery periods to assess the impact of music on vagal withdrawal and the recovery process, respectively. The last 5 minutes of the resting period and the total time exposure of the recovery period were considered for analysis; a strong digital filter was applied to smooth possible noise (detailed explanations regarding data analyses are better described below).
For the time domain, the root mean square of successive differences (RMSSD—vagal nerve index) and SD of R-R intervals (SDNN—global activity of autonomous system) were included (13). Fast Fourier transform was used to estimate the density of power specter, and this variable was quantified by 2 band pass: low frequency (LF)—0.04–0.15 Hz (global activity of autonomous systems) and high frequency (HF)—0.15–0.40 Hz (vagal nerve index) (13). Furthermore, a nonlinear analysis using a Poincaré plot was performed, presenting SD1 (parasympathetic) and SD2 (global activity) as extra indexes (14).
Heart Rate Variability: Functional Near-Infrared Spectroscopy
These same variables were obtained in the neuroimaging test to measure the influence of music on emotional responses through the autonomous system. The total music time was considered for HRV parameters, and a strong digital filter was applied to smooth possible noise. The HRV process was the same as described under the section Heart Rate Variability: Running Test.
Data Processing
A moving average window of 0.5 seconds was applied to each dependent variable of fNIRS and subsequently analyzed using software developed on Excel 2010 (Microsoft Office, Redmond, WA, USA) to stratify, account, share by voxels and PFC area/side (medial PFC [mPFC]—voxels 7, 8, 9, and 10; right dorsolateral PFC [RdlPFC]—voxels 11, 12, 13, 14, 15, and 16; left dorsolateral PFC [LdlPFC]—voxels 1, 2, 3, 4, 5, and 6), interpreting values to O2Hb and HHb. The difference between the neuroimaging test and baseline was represented in a.u.·μM−1. The difference between O2Hb and HHb was calculated previously (Diffbase) and during each track (Difftest). Finally, the difference of these differences (Difftest − Diffbase = activation) was considered as PFC activation (24), representing the increase of O2Hb, the decrease of HHb, or mutual considerations. The spectral and temporal components of HRV were analyzed with the Kubios HRV Software (Biosignal Laboratory, University of Kuopio, Finland).
Functional Near-Infrared Spectroscopy: Statistical Analysis
Data smoothing was performed through outlier identification, and subsequent replacement was performed applying multiple imputation in 6 cells for all variables. After descriptive statistics, data normality and homogeneity were checked using Shapiro-Wilk's test and Levene's test, respectively. The variables mPFC, LR, SAM (VAL and ACT), HF, LF, and SD1 presented Gaussian distribution and scalar dispersion (presented as mean ± SD), with possible parametric assessments (1-way ANOVA, followed by Bonferroni's post hoc test for multiple comparisons).
Nevertheless, RdlPFC, LdlPFC, and RMSSD did not show the same distribution (presented as median and interquartile range). In addition, the high dispersion does not allow any transformation (log10, square, etc.), enabling only nonparametric evaluations (Kruskal-Wallis' test followed by Mann-Whitney U-test when necessary). Furthermore, SDNN and SD2 initially had a nonparametric distribution, but the log10 transformations through computing variable were capable of correcting this dispersion, allowing parametric comparisons, as described above. The paired comparisons between the off period (PM condition) and music insertion were performed through paired Student's t-test for parametric variables (cerebral activation and HRV analysis) and Wilcoxon test for nonparametric variables. The significance level was set at p ≤ 0.05.
Five-Kilometer Run: Statistical Analyses
Data smoothing was performed through outlier identification, and subsequent replacement was performed applying multiple imputation in 8 cells for all variables. After descriptive statistics, data normality and homogeneity were checked using Shapiro-Wilk's test and Levene's test, respectively. Psychological variables (BRUMS, TRIMP, and FS) were processed through nonparametric analysis because of high dispersion (zero or negative values) and nonscalar distribution (presented as median and interquartile range). These variables were compared between PRE and POST moments using the Wilcoxon test and at each moment using the Kruskal-Wallis test followed by Mann-Whitney U-test when necessary.
HRV variables were analyzed through parametric methods (independent Student's t-test), only computing variables for RMSSD and SD1 (log10). This allowed a quick and standard correction. We have also conducted a temporal analysis for RMSSD, which was compared through repeated-measure ANOVA followed by Tukey's post hoc test (30-second windows). Rate of perceived exertion, HR, and LP were compared through repeated-measure ANOVA followed by Tukey's post hoc test for multiple comparisons (13 analyses), and the number of steps and total time were compared using 1-way ANOVA followed by Bonferroni's post hoc test. The significance level was set at p ≤ 0.05.
The smallest worthwhile change was used for total time in an attempt to check the possibility of the effect being prejudicial, trivial, or beneficial. The chances in the test were classified as <1% almost surely not, 1–5% very unlikely, 5–25% unlikely, 25–75% possible, 75–95% likely, 95–99% very likely, and >99% almost sure. When the chance of increase and decrease were both higher than 5%, the final examination was considered undetermined. The outcome description had the following criteria: first values mean the benefits, second values mean the trivial effect, and third values mean the prejudicial effect (first/second/third) (39).
Results
Functional Near-Infrared Spectroscopy
The 3 studied areas of the PFC were activated during music exposure without differences between the experimental conditions (Table 1): mPFC (F = 0.605; p = 0.615), RdlPFC (χ2 = 2.53; p = 0.471), and LdlPFC (χ2 = 0.67; p = 0.880). Nevertheless, the silence time applied after PM showed a significant increase in the medial and right dorsolateral areas of the PFC (residual effects during 5 off minutes) (mPFC: Z score Δ = −1.78, p = 0.042; RdlPFC: Z score Δ = −1.70, p = 0.049; LdlPFC: Δ = −1.53, p = 0.125). Also, Kruskal-Wallis' test did not shown significant differences between PFC regions compared between the experimental conditions (PM: χ2 = 0.53, p = 0.762; SM: χ2 = 0.42, p = 0.807; FM: χ2 = 1.94, p = 0.379; CS: χ2 = 0.36, p = 0.835).
Table 1: PFC (RdlPFC, mPFC, and LdlPFC) activation (a.u.·μM−1) for 3 different music tempi (PM, SM, FM, and CS), (n = 15 men).*
The chosen songs for PM, SM, and FM presented satisfactory enjoyment, pleasure, and activation levels (LR: 8.07 ± 1.87 a.u., VAL: 6.33 ± 1.62 a.u., ACT: 5.93 ± 1.02 a.u. for PM; LR: 7.60 ± 1.51 a.u., VAL: 6.67 ± 1.59 a.u., ACT: 5.47 ± 1.73 a.u. for SM; LR: 8.03 ± 1.33 a.u., VAL: 6.60 ± 1.18 a.u., ACT: 5.77 ± 1.54 a.u. for FM) without differences between them (LR: F = 0.39, p = 0.672; VAL: F = 0.20, p = 0.817; ACT: F = 1.67, p = 0.201). Nonetheless, CS showed significant differences compared with the other 3 for LR and ACT (LR: 5.53 ± 2.26 a.u., F = 6.56, p = 0.012; VAL: 5.60 ± 1.35 a.u., F = 1.64, p = 0.190; ACT: 3.23 ± 1.67 a.u., F = 3.13, p = 0.033).
The HRV analysis provided useful information regarding the emotional response to music. The following variables did not show any significant difference between the experimental conditions: SDNN: F = 2.14, p = 0.105; LF: F = 1.36, p = 0.262; HF: F = 1.52, p = 0.217; SD2: F = 2.26, p = 0.091. However, SD1 (general F = 2.95; p = 0.040) and RMSSD (general χ2 = 10.67; p = 0.14) demonstrated important and significant differences when CS (SD1: 35.33 ± 10.58 ms; RMSSD: 52 [20.1] ms) was compared with SM (SD1: 27.52 ± 7.76 ms, p = 0.014; RMSSD: 37 [12.2] ms, Z score Δ = −2.69, p = 0.006) and FM (SD1: 29.53 ± 9.55 ms, p = 0.048; RMSSD: 39 [9.4] ms, Z score Δ = −2.38, p = 0.017). Additionally, the paired comparisons between music exposure for PM and the off period (5 minutes of silence) did not show any differences for SDNN (t = −0.25; p = 0.801), RMSSD (Z = −0.56; p = 0.572), LF (t = −0.78; p = 0.443), HF (t = 0.62; p = 0.545), SD1 (t = 0.92; p = 0.369), and SD2 (t = −0.31; p = 0.358).
Psychological Analysis: Five-Kilometer Run
Analyses concerning personality were considered very consistent (low dispersion) (4 choleric, 3 melancholic, 1 phlegmatic, 7 sanguine). Eleven participants were classified as extroverted (choleric and sanguine) and 8 were considered unstable (choleric and melancholic). This feature might be influenced by the physical practice, which makes unlikely further stratifications.
The 6 BRUMS' domains demonstrated no differences between any experimental conditions regarding the PRE period (anger: χ2 = 2.61, p = 0.624; confusion: χ2 = 4.11, p = 0.391; depression: χ2 = 0.52, p = 0.901; fatigue: χ2 = 2.03, p = 0.730; tension: χ2 = 3.34, p = 0.501; vigor: χ2 = 0.98, p = 0.912) or POST period (anger: χ2 = 2.09, p = 0.718; confusion: χ2 = 3.27, p = 0.513; depression: χ2 = 0.77, p = 0.942; fatigue: χ2 = 1.43, p = 0.838; tension: χ2 = 5.70, p = 0.222; vigor: χ2 = 0.76, p = 0.943). Therefore, music exposure was not capable of modulating mood after 5 km of running, as presented in Figure 4. The general perception of effort (CR10) demonstrated no differences according to the experimental conditions after 5 km of running, which indicates a similar psychological stress (p > 0.05).
Figure 4: Mood variation (6 humor state domains) after 5 km of running, presented in 5 experimental conditions. No differences were found (p > 0.05). BRUMS = Brunel Mood Scale; CO = control condition (no intervention); PM = motivational songs (applied before 5 km of running); SM = slow motivational songs (applied during 5 km of running); FM = fast motivational songs (applied during 5 km of running); CS: calm songs (applied after 5 km of running).
Heart Rate Variability: Five-Kilometer Run
The general analysis of whole resting (before) and recovery (after) periods has shown a significant difference for PM and CS according to some parasympathetic indices. The music application before running did not affect the general RMSSD (CO: 63.1 ± 14.81 ms vs. PM: 54.5 ± 21.82 ms; t = 1.60; p = 0.120), SDNN (CO: 79.14 ± 22.63 ms vs. PM: 81.17 ± 22.12 ms; t = −0.24; p = 0.805), LF (CO: 61.65 ± 15.60 n.u. vs. PM: 61.52 ± 14.31 n.u.; t = 0.02; p = 0.981), HF (CO: 38.41 ± 15.59 n.u. vs. PM: 38.38 ± 14.31 n.u.; t = −0.01; p = 0.990), SD2 (CO: 101.14 ± 28.56 ms. vs. PM: 106.63 ± 28.50 ms; t = −0.52; p = 0.603), but this was capable of influencing the nonlinear index of vagal nerve SD1 (CO: 41.97 ± 9.61 ms vs. PM: 38.52 ± 15.58 ms; t = 19.32; p = 0.002). Moreover, temporal analysis of RMSSD according to motivational music application presented very interesting outcomes (less parasympathetic values) as shown in Figure 5.
Figure 5: Temporal analysis during resting period for root mean square of standard deviation (RMSSD: parasympathetic index) compared between 2 experimental conditions (CO = control condition, without intervention; PM = motivational songs, applied before 5 km of running). *Significant differences (p ≤ 0.05).
During recovery, calm songs not only demonstrated an effect for SDNN (CO: 50.96 ± 16.60 ms vs. CS: 45.42 ± 13.93 ms; t = −0.99; p = 0.331) and SD2 (CO: 67.58 ± 28.70 ms. vs. CS: 59.54 ± 24.25 ms; t = −0.82; p = 0.414) but also for RMSSD (CO: 4.78 ± 1.87 ms vs. CS: 7.69 ± 4.60 ms; t = 12.1; p < 0.001), LF (CO: 88.93 ± 5.55 n.u. vs. CS: 82.24 ± 10.11 n.u.; t = −2.22; p = 0.034), HF (CO: 11.09 ± 5.46 n.u. vs. CS: 17.75 ± 8.12 n.u.; t = 2.21; p = 0.035), SD1 (CO: 3.4 ± 1.3 ms vs. CS: 5.36 ± 3.22 ms; t = 11.86; p < 0.001). Similarly, a temporal analysis of RMSSD after running showed a visual distribution of vagal activation through 2 experimental conditions (higher parasympathetic values) (CO and CS—Figure 6).
Figure 6: Temporal analysis during recovery period for root mean square of standard deviation (RMSSD: parasympathetic index) compared between 2 experimental conditions (CO = control condition, without intervention; CS = calm songs, applied after 5 km of running). *Significant differences (p ≤ 0.05).
Psychophysiological and Performance Responses: Five-Kilometer Run
No HR or RPE differences were found for any protocol during all lap assessments. This represents a similar psychophysiological stress when analyzed in an isolated way. Nonetheless, the first 2 laps were dissimilar for SM and FM compared with the other 3 (CO, PM, and CS).
In addition, the final performance was considered the same despite the visual differences (CO: 1639 ± 143 seconds; PM: 1605 ± 148 seconds; SM: 1560 ± 146 seconds; FM: 1565 ± 144 seconds; CS: 1636 ± 150 seconds; F = 0.415; p = 0.797). Posterior analyses involving SWC showed a higher probability to achieve benefit by applying music as an ergogenic aid for 5 km of running (PM vs. CO: 39/60/2—possible; SM vs. CO: 89/11/0—likely; FM vs. CO: 85/15/0—likely). Second, the number of steps was considered similar (CO: 4543 ± 668 steps; PM: 4491 ± 584 steps; SM: 4476 ± 577 steps; FM: 4510 ± 605; CS: 4550 ± 647 steps; F = 0.041; p = 0.977). The psychophysiological comparisons for 5 km of running are demonstrated in Figure 7.
Figure 7: Psychophysiological (RPE and HR) and performance parameters (LP) during 5 km of running compared among 5 experimental conditions. RPE = rate of perceived exertion (15 points); a.u. = arbitrary units; HR = heart rate; PL = lap time; L = final lap; CO = control condition (no intervention); PM = motivational songs (applied before 5 km of running); SM = slow motivational songs (applied during 5 km of running); FM = fast motivational songs (applied during 5 km of running); CS = calm songs (applied after 5 km of running). *Statistical differences between SM and FM compared with CO, PM, and CS (p ≤ 0.05).
Discussion
The purpose of this study was to investigate how music could aid in 5 km of running (psychophysiological, performance, and recovery parameters), applied at different times, and then verify the cerebral and emotional response to such external stimulus. Thus, music was capable of activating the PFC area (medial and dorsolateral areas—left and right), evoking emotional responses (pleasure and activation), decreasing vagal tonus before running (as an indicator of preparation), increasing the same parasympathetic index after running (accelerating recovery process), improving initial speed up to 800 m, and providing a likely chance to achieve a better performance. Furthermore, it is important to note that this study accomplished a very real training situation (5 km of running—open space and self-selected songs), making the present findings useful for further applications.
Initially, music demonstrated an interesting activation in the 3 studied PFC areas (mPFC, RdlPFC, and LdlPFC), without differences between the experimental conditions. The PFC has been widely studied because of its great neural network with parallel areas (amygdala, hippocampus, cingulate cortex, and parietal cortex), representing an important link for external stimuli analysis. These cited areas indicate that chosen songs in some manner were capable of inducing emotional or memory responses. Also, the cingulate cortex and parietal cortex have been known for extensive performance monitoring and higher order sensory processing, respectively (30,40). However, such an emotional link was expected through music exposure, but residual effects with higher activation at 5 minutes during the silence period showed that music may induce at least medium-term benefits before physical tasks. These data agree with León-Carrión et al. (23) who tested a very similar task with 30 healthy subjects (15 men; 25.84 ± 7.62 years), applying a variety of film clips carrying emotional features (17 in.) and demonstrated higher HbO2 during stimuli cessation (off period).
Furthermore, SAM and HRV analyses provided an accompanying result. Self-Assessment Manikin has shown that CS conditions were considered calm compared with the other 3 conditions (PM, SM, and FM), and the RMSSD and SD1 (parasympathetic indices) presented higher values for CS compared with SM and FM. These results suggest an intriguing effect of music in modulating the vagal nerve through unconscious and conscious cortical areas in humans, decreasing stress through higher pituitary action (↓ interleukin-6, ↓ epinephrine, ↓ dehydroepiandrosterone) (7).
The music was not considered as a sufficient stimulus to alter psychological states due to the high intensity of the exercise. These outcomes contrast with other important studies involving open-loop exercises and probably this difference may indicate a more prominent psychological effect of music exposure for free exercises compared with time-trial protocols (38), which could create a previous calculus inside trained minds, inserting these participants in a parallel world of closed-loop exercise. As a result of this, the final outcome of this technique would be blocked from external influences (31,35). The first 800 m were markedly faster for SM and FM. Thereafter, similar responses were evident among all experimental conditions. Theoretically, the auditory stimulus is expected to prolong time to exhaustion by acting to limit the processing of fatigue-related cues through the efferent nervous system. Initially, participants were affected by music because they needed a time period to process all afferent information regarding peripheral receptors. As soon as the brain realized the exercise intensity, a mechanism called attentional switching occurred by directing attention to the most important signals (15,33). It decreased the effects of music on running performance because they have focused on internal processes instead of external influences.
This study has shown a real emotional response through music exposure before and after 5 km of running. The ability to increase arousal and to calm down was clearly demonstrated. Such an effect represents a very significant tool for arousal and recovery control without side effects. Several studies have applied music at previous times (during warm-up), finding controversial outcomes. This is because of the different methods applied (exercise models) (11,16), which must be taken into account for final recommendations.
Additionally, improvements in recovery using music are really rare (9), and the parasympathetic reduction after high-intensity cyclic exercise (maximum aerobic; ∼30 minutes), represents a simple and effective strategy for HRV manipulation through the sensory system. Also, the chosen songs were accepted by runners, and all components of music (tempo, melody, harmony, and style) were probably able to affect the arousal state through Conrad's proposed mechanism (7), decreasing physiological responses through arousal control.
Despite HR and RPE stabilization, the first and second lap showed significant improvements for music conditions SM and FM, and this represents an energy economy as shown previously by several studies (2,26). Despite PM indicating only a 39% chance of benefit and diminished vagal tonus for running, no differences were found for any variables with this experimental condition. It is considered likely that such a previous strategy was quickly covered by the physical exercise (more important peripheral cues for the central nervous system rather than residual effects of music). The same purpose may have happened in medium and final laps for SM and FM because of the curve distribution (Figure 7). Interestingly, SM and FM have very similar features, and despite asynchronous songs being tested, synchronous mechanisms could be aimed by runners unconsciously (20), making SM and FM more similar than expected.
A series of limitations in this study exist because of the research model (field applied research—outdoor). First of all, the weather represents one of the most important problems of this study because it is impossible to control. Second, asynchronous music can induce lesser outcomes than synchronous, but this strategy can fit better to real life (amateur runners using music for training). Third, the lap count may have created a kind of distraction factor (mainly for music conditions), making exercise more pleasurable, but losing knowledge of the end point. Finally, sedative songs (CS) did not follow the self-selected method applied for the other conditions (PM, SM, and FM). As a result of this, current outcomes regarding CS might be different according to varied cultures.
We conclude that music conditions are capable of activating the PFC (medial and dorsolateral—left and right), causing greater residual effects, increasing vagal withdrawal (HRV analysis) before running, improving parasympathetic recovery (HRV analysis) after running, and decreasing initial lap times (first and second laps).
Practical Applications
Enhancing performance or accelerating recovery is a common goal for coaches and athletes. Our study has demonstrated an interesting application of self-selected music for running, which is considered a simple method to achieve a variety of improvements. First of all, coaches and athletes are able to use music during training session as a means by which to improve running performance. During high-load microcycles when the necessity of motivation and enthusiasm is high, music-related interventions may aid runners to accomplish better results. Based on the current outcome, runners can also use music before competitions with a moderate chance of 39% to achieve benefits by applying music as an ergogenic aid. Finally, exercisers can use sedative auditory stimuli to accelerate fast-recovery (10 minutes) and prevent cardiac-related complications. For further replications, many aspects must be taken into consideration, including indoor track assessments (weather control), whole superficial cerebral analysis (fMRI or EEG), even during exercise (12), and synchronous music using online software to achieve real synchronization during cyclic exercises (39).
Acknowledgments
The authors thank the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) for graduate scholarships and Professor Jim Waterhouse for his meaningful contributions.
References
1. Aluja A, Garci[Combining Acute Accent]a Ó, Garci[Combining Acute Accent]a L. A psychometric analysis of the revised Eysenck Personality Questionnaire short scale. Pers Individ Dif 35: 449–460, 2003.
2. Atkinson G, Wilson D, Eubank M. Effects of music on work-rate distribution during a cycling time trial. Int J Sports Med 25: 611–615, 2004.
3. Biagini M, Brown L, Coburn J, Judelson D, Statler T, Bottaro M, Tran TT, Longo NA. Effects of self-selected music on strength, explosiveness, and mood. J Strength Cond Res 26: 1934–1938, 2012.
4. Bigliassi M, Estanislau C, Carneiro JG, Kanthack TFD, Altimari LR. Music: A psychophysiological aid to physical
exercise and sport. Archivos de Medicina del deporte 30: 311–320, 2103.
5. Borg G. Psychophysical bases of perceived exertion. Med Sci Sports Exerc 14: 377–381, 1982.
6. Bradley M, Lang P. Measuring emotion: The self-assessment manikin and the semantic differential. J Behav Ther Exp Psychiatry 25: 49–59, 1994.
7. Conrad C, Niess H, Jauch K-W, Bruns CJ, Hartl WH, Welker L. Overture for growth hormone: Requiem for interleukin-6? Crit Care Med 35: 2709–2713, 2007.
8. Crust L. Effects of familiar and unfamiliar asynchronous music on treadmill walking endurance. Percept Mot Skills 99: 361–368, 2004.
9. Eliakim M, Bodner E, Meckel Y, Nemet D, Eliakim A. Effect of rhythm on the recovery from intense
exercise. J Strength Cond Res 27: 1019–1024, 2013.
10. Eliakim M, Meckel Y, Gotlieb R. Motivational music and repeated sprint ability in junior basketball players. Acta Kinesiol Univ Tartu 18: 29–38, 2012.
11. Eliakim M, Meckel Y, Nemet D, Eliakim A. The effect of music during warm-up on consecutive anaerobic performance in elite adolescent volleyball players. Int J Sports Med 28: 321–325, 2007.
12. Fontes E, Okano A, De Guio F, Schabort E, Min L, Basset F, Stein DJ, Noakes TD. Brain activity and perceived exertion during cycling
exercise: An fMRI study. Br J Sports Med 2013. Epub ahead of print.
13. Gamelin F, Brethoin S, Bosquet L. Validity of the polar S810 heart rate monitor to measure RR intervals at rest. Med Sci Sports Exerc 38: 887–893, 2006.
14. Guzik P, Piskorski J, Krauze T, Schneider R, Wesseling K, Wykretowicz A, Wysocki H. Correlations between the Poincaré plot and conventional heart rate variability parameters assessed during paced breathing. J Physiol Sci 57: 63–71, 2007.
15. Hutchinson JC, Karageorghis CI. Moderating influence of dominant attentional style and
exercise intensity on responses to asynchronous music. J Sport Exerc Psychol 35: 625–643, 2013.
16. Jarraya M, Chtourou H. The effects of music on high-intensity short-term
exercise in well trained athletes. Asian J Sports Med 3: 233–238, 2012.
17. Karageorghis C, Mouzourides D, Priest D, Sasso T, Morrish D, Walley C. Psychophysical and ergogenic effects of synchronous music during treadmill walking. J Sport Exerc Psychol 31: 18–36, 2009.
18. Karageorghis C, Priest D. Music in the
exercise domain: A review and synthesis (Part I). Int Rev Sport Exerc Psychol 5: 44–66, 2012.
19. Karageorghis C, Priest D. Music in the
exercise domain: A review and synthesis (Part II). Int Rev Sport Exerc Psychol 5: 67–84, 2012.
20. Karageorghis CI, Terry PC, Lane AM, Bishop DT, Priest D-L. The BASES Expert Statement on use of music in
exercise. J Sports Sci 30: 953–956, 2012.
21. Kemi O, Wisløff U. High-intensity aerobic
exercise training improves the heart in health and disease. J Cardiopulm Rehabil Prev 30: 2–11, 2010.
22. Lan M, Lane A, Roy J, Hanin N. Validity of the Brunel Mood Scale for use with Malaysian athletes. J Sports Sci Med 11: 131–135, 2012.
23. Leon-Carrion J, Damas J, Izzetoglu K, Pourrezai K, Martín-Rodríguez J, Barroso y Martin J, Dominguez-Morales MR. Differential time course and intensity of PFC activation for men and women in response to emotional stimuli: A functional near-infrared spectroscopy (fNIRS) study. Neurosci Lett 403: 90–95, 2006.
24. León-Carrión J, León-Domínguez U. Functional near-infrared spectroscopy (fNIRS): Principles and neuroscientific applications. In: Neuroimaging-Methods. Bright P., ed. Rijeka: Intech, 2013. pp. 47–74.
25. León-Carrión J, Martín-Rodríguez J, Damas-López J, Pourrezai K, Izzetoglu K, Barroso y Martin J, Dominguez-Morales MR. A lasting post-stimulus activation on dorsolateral prefrontal cortex is produced when processing valence and arousal in visual affective stimuli. Neurosci Lett 422: 147–152, 2007.
26. Lim HBT, Atkinson G, Karageorghis CI, Eubank MR, Eubank MM. Effects of differentiated music on cycling time trial. Int J Sports Med 30: 435–442, 2009.
27. Marcora SM, Staiano W. The limit to
exercise tolerance in humans: Mind over muscle? Eur J Appl Physiol 109: 763–770, 2010.
28. Moghimi S, Kushki A, Guerguerian A, Chau T. Characterizing emotional response to music in the prefrontal cortex using near infrared spectroscopy. Neurosci Lett 525: 7–11, 2012.
29. Montinaro A. The musical brain: Myth and science. World Neurosurg 73: 442–453, 2010.
30. Murray E, O'Doherty J, Schoenbaum G. What we know and do not know about the functions of the orbitofrontal cortex after 20 years of cross-species studies. J Neurosci 27: 8166–8169, 2007.
31. Noakes T. Time to move beyond a brainless
exercise physiology: The evidence for complex regulation of human
exercise performance. Appl Physiol Nutr Metab 36: 23–35, 2011.
32. Rejeski W. The perception of exertion: A social psychophysiological integration. J Sport Psychol 3: 305–320, 1981.
33. Rejeski W. Perceived exertion: An active or passive process? J Sport Psychol 7: 371–378, 1985.
34. Skof B, Strojnik V. The effect of two warm-up protocols on some biomechanical parameters of the neuromuscular system of middle distance runners. J Strength Cond Res 21: 394–399, 2007.
35. St Clair Gibson A, Lambert EV, Rauch LHG, Tucker R, Baden DA, Foster C, Noakes TD. The role of information processing between the brain and peripheral physiological systems in pacing and perception of effort. Sports Med 36: 705–722, 2006.
36. Szmedra L, Bacharach DW. Effect of music on perceived exertion, plasma lactate, norepinephrine and cardiovascular hemodynamics during treadmill running. Int J Sports Med 19: 32–37, 1998.
37. Tate A, Gennings C, Hoffman R, Strittmatter A, Retchin S. Effects of bone-conducted music on swimming performance. J Strength Cond Res 26: 982–988, 2012.
38. Terry PC, Karageorghis CI, Saha AM, D'Auria S. Effects of synchronous music on treadmill running among elite triathletes. J Sci Med Sport 15: 52–57, 2012.
39. Vlist B, Bartneck C, Mäueler S. moBeat: Using interactive music to guide and motivate users during aerobic exercising. Appl Psychophysiol Biofeedback 36: 135–145, 2011.
40. Wood J, Grafman J. Human prefrontal cortex: Processing and representational perspectives. Nat Rev Neurosci 4: 139–147, 2003.
41. Yamasaki A, Booker A, Kapur V, Tilt A, Niess H, Lillemoe KD, Warshaw AL, Conrad C. The impact of music on metabolism. Nutrition 28: 1075–1080, 2012.
42. Yamashita S, Iwai K, Akimoto T. Effects of music during
exercise on RPE, heart rate and the autonomic nervous system. J Sports Med Phys Fitness 46: 425–430, 2006.
