Lower thresholds and stronger adaptation to pain in musicians reflect occupational‐specific adaptations to contact heat stimulation

Abstract

Background

Extensive audio‐motor training and psychological stress can cause professional musicians acute overstrain‐injury and chronic pain, resulting in damaged careers and diminished quality of life. It has also been previously shown that musicians might perceive pain differently than non‐musicians. Therefore, the aim of our study was to quantify differences between musicians and non‐musicians regarding their subjective responses to painful contact heat stimuli and assess how emotional traits might influence these responses.

Methods

Upon completing the StateTrait‐Anxiety‐Depression Inventory, 15 healthy musicians and 15 healthy non‐musicians from German universities received 15 noxious contact heat stimuli at the dorsal side of each hand and foot. After each stimulation, participants were asked to provide a pain rating from 0 to 10.

Results

Musicians not only reported significantly higher pain ratings after the first stimulation but also showed a significantly higher degree of habituation compared to non‐musicians. Additionally, musicians showed a significantly less pronounced difference regarding the pain rating of the hands compared to the feet than non‐musicians. Trait anxiety and trait depression scores had no effect on the pain rating or the habituation.

Conclusion

The more pronounced habituation of musicians might hint at a neuroplastic nociceptive alteration in musicians. The lack of significance between the psychological traits and their effect on the pain ratings is surprising but could be a result of both participant groups having stressful careers.

Significance

The findings of this report justify musicians' repetitive sensorimotor training as an important model for plasticity and contribute to a better understanding of pain perception in musicians.

1. INTRODUCTION

Nearly 86% of professional musicians report playing‐related musculoskeletal disorders (PRMD) while mastering music, potentially affecting their careers and well‐being (Kochem & Silva, 2018; Silva et al., 2015; Steinmetz et al., 2015). Despite this, pain research in musicians is scarce. However, one study shows that healthy, professional classical musicians had more tactile and pain sensitivity to heat and pressure in their fingers than healthy non‐musicians, similar to a non‐musician group with chronic pain, suggesting that musicians may perceive pain similarly to chronic pain patients (Zamorano et al., 2015). They inferred that musicians experience pain differently due to their enhanced sensorimotor areas from extensive practising (Zamorano et al., 2023). Musicians' perspectives on pain and their psychological traits may play a major role in pain perception, due to several links shown between mood disorders and pain (Kenny & Ackermann, 2015; Michaelides & Zis, 2019). Loveday et al. (2023) found that over half of a population of 254 global musicians reported high levels of anxiety and nearly a third reported having depression, arguing that their stressful environments and strong identification with their careers may influence their emotional states and perspectives. Therefore, investigating the differences in pain perception between musicians and non‐musicians would provide insights for better PRMD prevention.

This report is part of a preliminary study aimed at quantifying the subjective pain responses and evoked cortical potentials from rapid contact heat stimulation in musicians and non‐musicians. To explore whether the established altered pain perception in musicians is exclusive to their hands, both hands and feet were stimulated. This study additionally prepares for a larger study including dancers, who might also show training induced altered pain perception at the feet. The responses to rapid contact heat have not been explored in musicians but produce reliable subjective pain reports and nociceptive EEG responses (De Schoenmacker, Archibald, et al., 2022; Jutzeler et al., 2016; Rosner et al., 2018). The focus of this report was to quantify and discuss only the subjective pain perception in the hands and feet of musicians compared to non‐musicians. The goals were to examine the subjective pain responses to contact heat stimuli in musicians and non‐musicians and investigate the influence of the expressed emotional traits of anxiety and depression on pain perception. A secondary goal was to examine the change in pain ratings over time and whether musicians reported less pain to the consecutive heat stimuli differently than non‐musicians, since pain while playing is relevant for musicians.

Based on previous psychological and pain studies, we expected musicians to report higher pain ratings to all stimuli, have higher anxiety and depression scores compared to non‐musicians, and that these scores show an effect on the reported pain generally. We also expected both groups, albeit different, to show short‐term subjective habituation to the painful stimuli regardless of location, similar to previous contact heat studies (De Schoenmacker, Leu, et al., 2022; Paul et al., 2021). This research will provide great insight into musicians' pain perception and will contribute to the understanding and prevention of PRMD.

2. MATERIALS AND METHODS

2.1. Participants

Due to the premise of being a part of an exploratory preliminary study for a larger study, a total of 30 participants were recruited, where 15 participants were music performance students from Hannover University of Music, Drama and Media (HTMTH) majoring in piano, cello, violin, viola, trumpet and harp and the other 15 were non‐music students from the University of Veterinary Medicine Hannover and the Hannover Medical School. Each music performance student had at least 10,000 h of cumulative practice time in their life and had passed the extremely competitive entrance examination at the HTMTH. None of the participants were diagnosed with acute and chronic pain and/or neurological or psychiatric conditions. Additional criteria for exclusion were pregnancy and the intake of pain medication on a regular basis and/or 24 h before the experiment.

All participants gave written informed consent prior to participation. The study was conducted at the Institute of Music Physiology and Musicians' Medicine Hannover (IMMM) between October 2022 and March 2023. It was funded as part of a doctoral position at the IMMM and approved by the ethics committee of the Hannover Medical School (Reference: 10328_BO_S202). All experimental procedures abided by the ethical standards of the 1964 Declaration of Helsinki. This study did not include clinical patients.

2.2. Study design and participant tasks

The stimulation paradigm is based on studies by Jutzeler et al. (2016) and Rosner et al. (2018). Participants received 15 contact heat stimuli at the dorsal side of their left and right hands (dermatome C6) and feet (dermatome L5), where each stimulus had a starting baseline temperature of 42°C and a peak temperature of 52°C. This resulted in four different stimulation periods and locations for a total of 60 stimuli. The painful contact heat stimuli were applied via a contact plate thermode (5.76 cm²) from the Thermosensory Stimulator Advanced 2 stimulation device (TSA 2, MEDOC, Israel) and the stimulation protocol was controlled by the Medoc Main Station software (Version Arbel 7.0.0.17). The thermode is capable of rapid heating and cooling rates of up to 70°C/s and 40°C/s respectively, according to the manufacturer. Upon reaching the peak temperature, the contact plate was immediately and automatically cooled back down to the baseline temperature with the maximal cooling rate. The inter‐stimulus interval was randomized between 8 and 12 s. After each individual stimulation, the position of the thermode was adjusted slightly to reduce peripheral adaptation (Greffrath et al., 2007). The 15 consecutive stimuli were applied at each location with only a short break between each location, and the order of stimulation at each location was performed in a randomized order for each participant.

At any point, if the stimulation was perceived as too painful, participants were allowed to either continue with stimuli that started with a lower baseline temperature of 35°C and/or quit the experiment entirely. Nevertheless, all participants received a second set of 15 stimuli at each hand and foot following the first 60 stimuli with a lower baseline temperature of 35°C. The order of the locations of stimulation for the stimuli with the lower baseline was also randomized for each participant separately than the previous set of stimuli. Since no participants aborted the stimuli with the increased baseline, and prior research shows that the increased baseline temperature stimuli produce robust pain ratings and nociceptive responses (Kramer et al., 2013), we excluded the pain ratings reported from the lower baseline stimuli for this report.

During the stimulation protocol, participants were instructed to sit as still and relaxed as possible in a chair while fixing their gaze on a dot at the level of their eyes. They were asked to rate the pain intensity of the stimuli 2 s after each stimulation based on the 0–10 numeric rating scale (NRS), with 0 representing ‘no pain’ and 10 representing ‘the worst pain imaginable’ (Breivik et al., 2008; Jensen et al., 1989). Prior to the full protocol, participants were given five test stimuli with the 42°C baseline temperature on the left hand to familiarize themselves with the stimuli and their tasks. The overall study design is summarized in Figure 1.

FIGURE 1.

Experimental design and parameters of the study. (a) Each participant completed the State Trait Anxiety Depression Inventory ‐ Trait (STADI‐T) questionnaire. (b) Stimulation sites at C6 and L5 on either side of the body. (c) Temperature curves of the contact heat stimuli with an increased baseline (42°C) followed by lower baseline (35°C) produced by TSA2 stimulation device with thermode. (d) Two seconds following each stimulation, participants provided their subjective pain rating with the numerical rating scale (NRS); The TSA2 and NRS images were borrowed from Medoc: Advanced Medical Systems and Physiopedia contributors (2022), respectively. The remaining images were created with BioRender.com.

2.3. Trait anxiety and depression questionnaire

After providing personal demographic information, all participants completed the German version of the reliable and validated State‐Trait‐Anxiety‐Depression Inventory regarding Trait Anxiety and Trait Depression (Laux et al., 2013). The STADI‐T assesses the participant's tendency to experience anxiety or depressiveness in their daily lives and consists of 20‐items that investigate trait anxiety and depression on a 4‐point Likert scale (1 = not at all, 4 = very much so). Both scores, Trait Anxiety and Trait Depression, were determined by assigned values based on the participant's sex and sum of the appropriate items for each category, and both had very good Cronbach reliability values of 0.88 and 0.89, respectively (Laux et al., 2013). The definitions and appropriate items for Trait Anxiety and Trait Depression are listed in Table S1 in the Supplementary Information.

2.4. Data analysis

Statistical analyses were conducted in RStudio (Version 4.3.0) with the package lme4 (Version 1.1–33). Statistical significance level was set to α = 0.05. Demographic data between the two groups were compared with an unpaired Student's t‐test and the age was normalized by min‐max scaling. The mean values and standard deviations of the STADI‐T scores were first determined and their distributions tested for normality via a Shapiro–Wilk's test before a pairwise comparison between groups was performed with an unpaired Student's t‐test. Then, the STADI‐T scores were normalized by min‐max scaling across all participants with respect to the mean score of Trait Anxiety and Trait Depression, respectively.

For each one of the 30 participants, we had 60 pain ratings (4 locations × 15 stimuli). Hence, we chose a hierarchical linear model (HLM) to make use of all the 1800 data points while considering the other variables embedded within the individuals. A linear relationship between the pain rating (R) and the stimulus (stim) was assumed to investigate the differences in pain rating and habituation between both groups. Furthermore, the influence of the parameters group (musician/non‐musician, G), stimulation location (hands/feet, L) and the normalized participants' age (A) on the initial pain rating were investigated. Additionally, the interactions of both group and habituation and group and location were included in the model. However, it should be noted that two separate HLMs were used with the previous variables to investigate the individual influence of the normalized Trait Anxiety scores (Anx) and the normalized Trait Depression scores (Dep) on the pain rating and the habituation. This first HLM considers the influence of the normalized trait anxiety score (anxiety model):

$$R1\sim \mathit{Anx}\times \mathit{\text{stim}}+G\times \mathit{\text{stim}}+G\times L+A+\left(1+\mathit{\text{stim}}\vee \mathit{ID}\right)$$

This second HLM considers the influence of the normalized trait depression score (depression model):

$$R2\sim \mathit{Dep}\times \mathit{\text{stim}}+G\times \mathit{\text{stim}}+G\times L+A+\left(1+\mathit{\text{stim}}\vee \mathit{ID}\right)$$

Random effects of the pain rating and its habituation with subsequent stimuli were allowed in the model by linking them to the participants' unique identification code (ID).

The residuals of the statistical models were tested for normality by visually inspecting the Q‐Q plot and plotting a histogram against a normal distribution of the same mean and variance. The effect of a certain variable on the habituation is given by the interaction between this variable and the stimulus number. Random effects on the subject level were included. The parameters group and location were dummy‐coded as dichotomous variables where G = 1 refers to musicians, G = −1 refers to non‐musicians and L = 1 refers to the hands, while L = −1 refers to the feet. For the HLMs, it is important to note that the left and right side of the body were not differentiated, that is, for the hands and feet both, the left and right side were included together.

3. RESULTS

3.1. Participant demographics

Participants' demographic data are summarized in Table 1. No participants were excluded from the analysis. Musicians were significantly younger than the non‐musicians (Table 1, p = 0.002), while showing no significant difference in height or sex distribution. Females were significantly younger in the musician group compared to non‐musicians (Table 1, p < 0.001). All but three participants were right‐handed, and the left‐handed musicians handled their instrument similarly to the right‐handed musicians. All musicians played instruments that required fine motor movements in their hands, and only one musician played a trumpet, compared to the others who played string instruments or the piano.

TABLE 1.

Demographics of participants.

Variables	Total (n = 30)	Musicians (n = 15)	Non‐musicians (n = 15)	t‐score	p value
Sex, female/male	18/12	9/6	9/6	‐	‐
Age, years, M ± SD	24.8 ± 4.3	22.5 ± 4.0	27.1 ± 3.3	3.47	0.002
Female	24.3 ± 4.3	21.3 ± 2.9	27.3 ± 3.3	3.73	<0.001
Male	25.5 ± 4.4	24.2 ± 5.1	26.8 ± 3.5	1.01	0.318
Height, cm, M ± SD	174.4 ± 8.8	172.6 ± 9.8	176.3 ± 7.6	1.14	0.262
Female	168.9 ± 5.8	166.6 ± 6.1	171.3 ± 4.7	1.80	0.082
Male	182.7 ± 5.4	181.7 ± 6.7	183.7 ± 4.1	0.613	0.544
Handedness, right/left	27/3	13/2	14/1	‐	‐
Right, female/male	15/12	7/6	8/6	‐	‐
Left, female/male	3/0	2/0	1/0	‐	‐

Note: The values for sex and handedness are the total number of participants from the respective group while the values for the age and height are the mean and standard deviations. The t‐scores and p values were taken between participant groups for the total age and height, as well as for the female and male participants within the groups. Musicians, especially females, were significantly younger than non‐musician controls.

Abbreviations: M, mean; SD, standard deviation.

3.2. STADI‐T scores

The STADI‐T scores for both groups are summarized in Table 2. Musicians had numerically, but not statistically significant, higher scores for both Trait Anxiety and Trait Depression, with a trend towards higher Trait Depression scores in musicians (Table 2, p = 0.07).

TABLE 2.

Questionnaire baseline characteristics.

Questionnaire	Sub‐score	Musicians	Non‐musicians	t‐score	p value
STADI‐T	Trait Anxiety	59.60 ± 9.63	57.27 ± 10.54	0.63	0.53
	Trait Depression	60.93 ± 5.15	57.80 ± 3.84	1.89	0.07

Note: Each value is the mean score or sub‐score for the appropriate group ± standard deviation, with the respective t‐score and p value taken between groups. There were no significant differences between groups for all scores, but a trend was seen for a higher Trait Depression score in musicians.

Abbreviation: STADI‐T, State–Trait‐Anxiety‐Depression Inventory.

3.3. Pain rating and pain habituation

The results and estimates (E) of the linear hierarchical models for the pain ratings are depicted in Table 3 for the anxiety model and in Table 4 for the depression model. For both models, the intercept of the pain rating was significantly different from 0 (Table 3: Intercept = 5.43, p < 0.001; Table 4: Intercept = 5.43, p < 0.001). There were also significant effects of consecutive stimuli on the pain rating in general (Table 3: stim = −0.11, p < 0.001: Table 4: stim = −0.11, p < 0.001), indicating that the pain rating significantly decreased during the period of 15 stimuli, regardless of group and stimulation location. When looking at the pain rating with respect to the group, regardless of the stimulation location, there were significant differences between musicians and non‐musicians regarding the pain rating (Table 3: G = 1.03, p = 0.008; Table 4: G = 1.06, p = 0.007), as well as a significant interaction of the number of the stimulus (i.e. habituation) and the group (Table 3: stim × G = −0.03, p = 0.011; Table 4: stim × G = −0.04, p = 0.013). Likewise, there was a significant effect of the location of the stimulation (Table 3: L = −0.16, p < 0.001; Table 4: L = −0.16, p < 0.001) as well as a significant interaction between group and stimulation location (Table 3: G × L = 0.08, p = 0.013; Table 4: G x L = 0.08, p = 0.013). Conversely, Trait Anxiety and Trait Depression scores and the participants' age had no significant effect on pain rating or habituation, where appropriate (Table 3 & Table 4: p > 0.05).

TABLE 3.

Estimations of predictors and random effects of the hierarchical linear model for the pain ratings regarding the normalized trait anxiety scores.

NRS Predictors	Estimate (E)	Confidence interval	p value
Intercept	5.43	4.79 to 6.07	< 0.001
Trait anxiety score (T Anx)	0.06	−0.62 to 0.73	0.868
Stimulus (stim)	−0.11	−0.14 to −0.08	<0.001
Participant group (G)	1.03	0.27 to 1.80	0.008
Stimulation location (L)	−0.16	−0.22 to −0.09	<0.001
Participant age (A)	0.26	−0.53 to 1.05	0.524
stim × T Anx	0.02	−0.00 to 0.05	0.101
stim × G	−0.03	−0.06 to −0.01	0.011
G × L	0.08	0.02 to 0.15	0.013

Note: (Intercept) = first pain rating reported; The Trait Anxiety scores were normalized. The estimates (E) for the variables themselves describe their effect on the pain rating while the estimates of the interaction of the variable with ‘stim’ describe their effect on the habituation. The estimates for (Intercept), stim, G, L, stim × G and G × L are significant (α = 0.05). The total number of observations was 1800 across all 30 participants. The random effects of the model were normal: σ ² = 1.93, ${\tau }_{00\mathit{ID}}$ = 3.07, ${\tau }_{11\mathit{ID}.\mathit{\text{stim}}}$ = 0.00, ${\rho }_{01\mathit{ID}}$ = −0.13, ICC = 0.62, Marginal R ² = 0.14 and Conditional R ² = 0.67. The random effects are defined as the following: σ² = variance of the residuals; ${\tau }_{00\mathit{ID}}$ = variance of the intercept of the HLMs; ${\tau }_{11\mathit{ID}.\mathit{\text{stim}}}$ = variance of the slope of the HLMs; ${\rho }_{01\mathit{ID}}$ = correlation between stimulus and participant ID.

Abbreviation: NRS = Numeric Rating Scale; ICC, Intraclass correlation.

TABLE 4.

Estimations of predictors and random effects of the hierarchical linear model for the pain ratings regarding the normalized trait depression scores.

NRS Predictors	Estimate (E)	Confidence interval	p value
Intercept	5.43	4.79 to 6.06	< 0.001
Trait depression score (T Dep)	−0.23	−0.93 to 0.47	0.523
Stimulus (stim)	−0.11	−0.14 to −0.08	<0.001
Participant group (G)	1.06	0.30 to 1.83	0.007
Stimulation location (L)	−0.16	−0.22 to −0.09	<0.001
Participant age (A)	0.16	−0.62 to 0.93	0.693
stim × T Dep	0.01	−0.01 to 0.04	0.335
stim × G	−0.04	−0.07 to −0.01	0.013
G × L	0.08	0.02 to 0.15	0.013

Note: (Intercept) = first pain rating reported; The Trait Depression scores were normalized. The estimates (E) for the variables themselves describe their effect on the pain rating while the estimates of the interaction of the variable with ‘stim’ describe their effect on the habituation. The estimates for (Intercept), stim, G, L, stim x G and G x L are significant (α = 0.05). The total number of observations was 1800 across all 30 participants. The random effects of the model were normal: σ ² = 1.93, ${\tau }_{00\mathit{ID}}$ = 3.03, ${\tau }_{11\mathit{ID}.\mathit{\text{stim}}}$ = 0.00, ${\rho }_{01\mathit{ID}}$ = −0.09, ICC = 0.62, Marginal R² = 0.13 and Conditional R² = 0.67. The random effects are defined as the following: σ² = variance of the residuals; ${\tau }_{00\mathit{ID}}$ = variance of the intercept of the HLMs; ${\tau }_{11\mathit{ID}.\mathit{\text{stim}}}$ = variance of the slope of the HLMs; ${\rho }_{01\mathit{ID}}$ = correlation between stimulus and participant ID.

Abbreviation: NRS = Numeric Rating Scale; ICC = Intraclass correlation.

The pain ratings with respect to the stimulus number for both groups regardless of stimulation location are visualized as linear fits in Figure 2, which show that musicians report higher pain ratings and show a more pronounced habituation to painful stimuli compared to non‐musicians.

FIGURE 2.

Linear fits for all pain ratings in musicians and non‐musicians. Musicians (orange) show higher pain ratings (0 to 10 NRS; 900 pain ratings per group) and steeper habituation com‐pared to non‐musicians (blue). Each pain rating is represented by a dot in the respective color. The course of the pain rating across the 15 stimuli is depicted as a line in the respective color, where the first of the 15 stimuli is marked as #0.

4. DISCUSSION

We investigated differences between musicians and non‐musicians regarding their subjective responses to consecutive painful stimuli, with a primary focus of the influence of emotional traits on the pain ratings, and a secondary focus on the temporal evolution of the pain responses in general (i.e. habituation). We found that musicians perceived the stimuli as more painful but interestingly had a stronger habituation. We also found that in both groups, participants reported lower pain ratings for the hands as compared to the feet. This tendency, however, was smaller for the musicians than for the non‐musicians. There was no significant influence of emotional traits on pain or habituation.

4.1. Effect of psychological traits

We expected Trait Anxiety and Trait Depression to have an influence on pain, as it was shown that in patients, acute pain and mood disorders are risk factors for one another (Michaelides & Zis, 2019). However, no significant effect was found, which could be explained by our relatively small sample size. Another explanation may be that participants had neither a clinical, psychological, nor pain diagnosis and that in both groups the anxiety and depression scores merely showed a tendency towards either condition. Since participants were pain‐free, there might not be as strong of a relationship between pain and psychological conditions in our sample as in clinically diagnosed patients.

Based on previous reports of anxiety and depression in musicians due to their stressful environments (Loveday et al., 2023; Vaag et al., 2015, 2021), we expected a significant difference between both groups regarding Trait Anxiety and Depression scores. We found no significant group differences, however, both groups reported high scores and thus its effect on the pain rating cannot be resolved. Indeed, the non‐musicians mainly consisted of graduate students from veterinary and medical universities, stressful academic environments, which might account for the lack of a significant group difference. Nevertheless, musicians still displayed slightly higher, albeit not significant, Trait Anxiety and Depression scores, which should be amplified with a larger sample size.

4.2. Musicians: Higher pain, smaller hand‐feet difference and stronger habituation

All participants reported higher pain ratings for their feet compared to their hands. The enhanced pain sensitivity of the feet aligns with a study by Ng et al. (2024) where they show that feet have better discrimination of painful stimuli than hands. The interaction of stimulation location and group from our models suggests that this discrepancy between hands and feet is significantly smaller for musicians and larger for non‐musicians. This finding corroborates Zamorano et al. (2015), who found that musicians perceived heat stimuli to their fingers as more painful than non‐musicians. They also suggested that this enhanced sensitivity in musicians' hands results from prolonged and intense training, which may have induced similar enhanced pain perception as those with chronic pain (Zamorano et al., 2015, 2017).

In fact, in our population, musicians consistently reported more pain, regardless of stimulus location. We suspect that this might be related to the musicians' enhanced attention to somatosensory input (Zamorano et al., 2015). Likewise, we also suggest that the higher pain in musicians could be due to playing‐induced neuroplastic alterations within the somatosensory domain. However, previous findings showed that musicians have complex relationships with pain as it affects their careers and as their personalities are closely linked with their profession (Loveday et al., 2023; Vaag et al., 2015, 2021). Interestingly though, a study by Korte et al. (2023) showed that music students tend to catastrophize pain less severely compared to non‐music students. These studies suggest that increased pain in musicians can be explained by a combination of physiological and complex psychological influence factors, and this relationship between pain attitudes and pain needs to be examined more closely.

Nevertheless, musicians in this study showed a significantly stronger short‐term habituation to pain than non‐musicians. This cannot be explained by a numerical bias, as musicians not only reported significantly higher pain ratings in the beginning but also reported zero pain for later stimuli more often than non‐musicians. This interesting finding aligns with Zamorano et al. (2019, 2023), who suggests musicians have an enhanced top‐down pain modulation and reduced pain related interference in daily activities, despite their enhanced pain sensitivity. A recent contact heat evoked potential (CHEP) study by De Schoenmacker et al. (2024) suggests the preservation of subjective and objective short‐term habituation in chronic pain patients. The reduction of the response to a painful stimulus can be influenced by both the central and the peripheral nervous system (Bingel et al., 2007; Greffrath et al., 2007; Rennefeld et al., 2010). Centrally, a reduced activation of the thalamus and increased activation of the anterior cingulate cortex may reflect habituation (Bingel et al., 2007). Zamorano et al. (2017, 2019) showed enhanced insular connectivity in musicians due to their sensorimotor training, which could explain the reported enhanced habituation to pain in musicians, as the insular cortex, together with the anterior cingulate cortex, has efferent control over the brainstem (Craig, 2003).

On the other hand, these fMRI studies did not control for pain‐independent somatosensory stimulation, suggesting a broader habituation network. Indeed, a controlled study by Paul et al. (2021) primarily showed involvement of the anterior mid‐cingulate cortex in pain‐specific habituation, which can be interpreted as a reduction in the affective‐motivational response to persistent nociceptive stimulation. In musicians, affective‐motivational networks are also trained and altered in addition to sensorimotor networks (Fasano et al., 2020; Matthews et al., 2020; McPherson et al., 2016; Miendlarzewska & Trost, 2014). Additionally, the insular cortex is well known for being involved in bodily awareness, which could evoke stronger short‐term habituation (van der Miesen et al., 2023) and could explain why musicians have a more pronounced habituation.

Finally, Greffrath et al. (2007) showed a clear peripheral involvement in the reduction of both subjective and objective responses to repeated contact heat stimuli. Since the alteration of peripheral aspects of pain transmission in musicians have not been investigated yet, more research is needed to determine the differences between musicians and non‐musicians regarding the peripheral nervous system that leads to the more pronounced habituation in musicians. Including pain threshold data and comparing the CHEPs between the hands and the feet might indicate which aspects of the nervous system might be predominantly involved in the group differences.

4.3. Limitations of the study

Our findings must be interpreted in the light of certain limitations. One main limitation is the small sample size. We could not make a reasonable power assumption, since this study was designed as an exploratory preliminary study with the aim of establishing a paradigm for a larger study, where the sample size was determined prior to be 22 participants in each group (musicians, dancers and non‐musicians/non‐dancers). Nonetheless, the groups were homogeneous and support the credibility of the results. Another limitation is the significant age difference between the groups, as age influences pain. However, we did not find a significant influence of the participants' age on pain or habituation either, indicating that these differences also cannot be explained by age. Jutzeler et al. (2016) assessed pain in various age groups and concluded an age‐range from 18 to 40 years to be homogenous, which might explain the lack of a significant age effect in our population.

We also did not investigate the influence of different instruments on pain. However, even though different instruments may have differing demands with regard to temporo‐spatial precision, musicians at this level have more than 10,000 h of practice (Ericsson et al., 1993) suggesting that these differences take place at the highest level of fine‐motor control. Therefore, the effect of performing with different instruments may be of interest when comparing musicians, but negligible in comparison to non‐musicians. Moreover, we did not distinguish the body‐sides; especially between left‐ and right‐handed participants, differences might occur.

Finally, the hierarchical linear models we used have limitations: They do not take into account that the pain rating is bound between 0 and 10, which might lead to inaccurate confidence intervals of the estimations. This could be avoided by a beta‐regression; however, the necessary transformations would make the estimates uninterpretable. Furthermore, habituation on the time scale of our study might not be linear, which our model assumes.

5. CONCLUSIONS

In summary, musicians perceived a significantly higher pain intensity to contact heat stimuli and at the same time habituated significantly more to the painful stimuli than non‐musicians. This difference is possibly due to neuroplastic alterations resulting from the refined motor movements and the intentional sensorimotor attention during musical training (Zamorano et al., 2019). Future studies should investigate to what degree this altered pain perception can be quantified by the cortical pain response (i.e. CHEPs) and how the musicians' complex emotions towards pain can influence their perception. This study further justifies musicians as an interesting group for pain research and contributes to understanding the characteristics of musician's pain processing and perception in hopes to reduce the incidence and prevalence of PRMD.

AUTHOR CONTRIBUTIONS

RD and FS conducted the experiments, analysed the data and wrote the original manuscript together. AL and MK organized the ethics and experiment proposal, organized the overall project and refined sections related to their expertise, while AL also helped with conducting some experiments. EA, AA and PR provided neurological contextualization and helped with the revision process.

FUNDING INFORMATION

This project would not have been possible without the generous funding from the Institute for Music Physiology and Musicians' Medicine (IMMM). In addition, RD and PR were supported by the MSc/PhD program ‘Neurosciences’—International Max Planck Research School at the Georg August University Göttingen while FS was supported by the IMMM.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

Supporting information

Table S1.

Divarco, R. , Sternkopf, F. , Karst, M. , Altenmüller, E. , Ramasawmy, P. , Antal, A. , & Lee, A. (2025). Lower thresholds and stronger adaptation to pain in musicians reflect occupational‐specific adaptations to contact heat stimulation. European Journal of Pain, 29, e4738. 10.1002/ejp.4738