The sample of this study consisted of 393 musicians including 23% professional and 49% non-professional orchestra musicians and 28% singers in amateur choirs (see Table 1). The classification into these musical subgroups was based on the musical level of the orchestras. Professional orchestra musicians had main occupations at radio symphony orchestras or philharmonic orchestras. The non-professional orchestra musicians came from student orchestras and semiprofessional orchestras. The amateur choir singers were amateur singers in semiprofessional choirs. The sample is a sub-sample of the study Spahn et al. (2021).

In total, 59.6% were female musicians. There was a significant distribution difference between the musical subgroups [χ² (398, 2) = 27.7, p < 0.001] with more females among the choir singers than among the orchestra musicians.

The mean age of the sample was 30.2 years [SD = 13.6 years]. The professional orchestra musicians were significantly older than the non-professional orchestra musicians and the amateur choir singers [F (2,381) = 74.5, p < 0.001], with the latter two being mostly students.

The distribution of the instruments were 45% strings (with 79% in the professional and 57% in the non-professional orchestras), 13% woodwind instruments (with 12% in the professional and 22% in the non-professional orchestras), 29% singers (all in the choirs) and 13% other instruments such as percussions and brass instruments.

After concerts of the orchestras and choirs, the musicians were asked to fill in the questionnaires as pen-and-paper version. The concerts were regular concerts of the music group in line with their normal performance schedule and with standard repertoire. In the off-stage facilities, the musicians were given the questionnaire by an experimenter, who informed them about the study and that participation was completely anonymous and voluntary. The study was granted ethical approval by the Ethics Committee of the University Clinic Freiburg.

The statistical analyses were performed using SPSS (Version 29, Armonk, NY: IBM Corp.). Descriptive statistics including mean and standard deviation (SD) were calculated for each variable. To analyse distribution differences of non-parametric variables Chi-square (χ²) tests were performed. Parametric comparisons on the PQM scales and the NEO-FFI personality traits were calculated with multivariate ANOVAs. On significance, post hoc analyses with Tukey HSD correction were performed. A canonical correlation analysis (CCA) was conducted using the NEO-FFI personality traits and the PQM scales to evaluate the multivariate shared relationship between the two variable sets. The level of statistical significance was set at 0.05.

The multivariate analysis of the NEO-FFI scales across the three musical subgroups within the whole sample found significant effects in the scales openness [F (2,390) = 3.38, p = 0.035] and conscientiousness [F (2,390) = 3.83, p = 0.022] (Table 2). The effect in openness was caused by higher values in the professional musicians compared to the choir singers [Post-Hoc, p = 0.035] and in conscientiousness by higher values in the professional musicians compared to the amateur musicians [Post-Hoc, p = 0.016].

There were also significant main effects for gender in the neuroticism scale [F (1,383) = 6.32, p = 0.001] and in the agreeableness scale [F (1,383) = 15.66, p < 0.001] with higher values for the female musicians in both scales. However, there were no significant interaction effects for gender and musical subgroup.

The mean values of the PQM scales for each musical subgroup are listed in Table 3. The multivariate analysis across the three musical subgroups found several significant main effects. Before the performance, in the scale symptoms of MPA [F (2,390) = 6.24, p = 0.002] the non-professional musicians were significantly higher than the professional musicians [Post-Hoc, p = 0.037] and the choir singers [Post-Hoc, p = 0.004] and in the self-efficacy scale [F (2,390) = 5.11, p = 0.006] the choir singers were significantly higher than the professional [Post-Hoc, p = 0.013] and the non-professional orchestra musicians [Post-Hoc, p = 0.016]. During the performance, in the scale functional coping [F (2,390) = 2.19, p = 0.02] the choir singers were significantly higher than the non-professional musicians [Post-Hoc, p = 0.016], in the scale symptoms of MPA [F (2,390) = 10.68, p < 0.001] the non-professional musicians were significantly higher than the choir singers [Post-Hoc, p < 0.001] and the professional musicians [Post-Hoc, p = 0.013] and in the self-efficacy scale [F (2,390) = 5.98, p = 0.003] the choir singers were significantly higher than the professional [Post-Hoc, p = 0.006] and the non-professional orchestra musicians [Post-Hoc, p = 0.008]. After the performance, in the scale functional coping [F (2,390) = 15.91, p < 0.001] the choir singers were significantly higher than the professional [Post-Hoc, p < 0.001] and the non-professional orchestra musicians [Post-Hoc, p < 0.001] and in the scale symptoms of MPA [F (2,390) = 7.83, p < 0.001] the choir singers were significantly lower than the professional [Post-Hoc, p < 0.001] and the non-professional orchestra musicians [Post-Hoc, p = 0.003].

The canonical correlation analysis (CCA) yielded five functions with 59.5% explained variance for the full model across all functions with the first three functions being significant. Since the first function [F (45,1,698,5) = 4.81, p < 0.001] already explained 55.4% of the variance shared between the variable sets, only this function was considered for further analysis. In this function, the questionnaires correlated with Pearson’s r = 0.498 with each other (see Figure 1). The explained variance for the NEO-FFI questionnaire was 32.7% and for the PQM 35.9%.

For the analysis of the individual predictor variables, only correlation coefficients above r = 0.5 were considered. The results showed that neuroticism, extraversion and conscientiousness were the primary contributors for the prediction on the NEO-FFI side and the self-efficacy scales at all performance times as well as all scales after the performance on the PQM side. While extraversion and conscientiousness were positively related, neuroticism was negatively correlated. For the PQM scales, self-efficacy and functional coping showed positive and symptoms of MPA negative correlation coefficients.

The correlations between the NEO-FFI personality traits with the self-assessed judgment score of the performance quality found significant correlation coefficients only for neuroticism [r = −0.22, p < 0.01] and extraversion [r = 0.23, p < 0.01]. These correlations are moderate to low. The negative correlation of neuroticism describes that musicians with higher neuroticism rated their performance lower. Musicians with higher values of extraversion rated their performance as better.

The types of MPA were distributed to 52% in Type 1, 23% in Type 2 and 24% in Type 3. In a multivariate analysis, extraversion, openness, neuroticism and conscientiousness showed significant main effects across the three types of MPA [F (2,390) > 7.0, p < 0.01] (see Figure 2).

The detailed comparisons showed that musicians of MPA Type 1 had significantly lower mean values in the neuroticism trait than the musicians of the other Types [Post-Hoc, p < 0.01]. Furthermore, the musicians of MPA Type 1 had also significantly higher values in the extraversion scale than the musicians of Type 2 [Post-Hoc, p < 0.01] but not of Type 3. Here, the musicians of Type 2 had significantly lower values than of Type 3 [Post-Hoc, p = 0.017]. For the openness, musicians in Type 1 were significantly higher compared to musicians in Type 2 [Post-Hoc, p = 0.002] and in Type 3 [Post-Hoc, p = 0.047]. Musicians of the MPA Type 1 were also higher in the conscientiousness trait than musicians of Type 2 [Post-Hoc, p < 0.01] and of Type 3 [Post-Hoc, p = 0.015]. Except for the extraversions scale, the mean values of the musicians in Type 2 and 3 did not differ significantly in the personality traits.

In the NEO-FFI personality traits, the three musical subgroups differed only in openness and conscientiousness. Professional orchestra musicians showed significantly higher openness and conscientiousness than non-professional orchestra musicians and amateur choir singers.

This is in contrast to the results of Kuckelkorn et al. (2021), where the other personality traits did also differ significantly between professional and non-professional musicians. Using the NEO-FFI personality trait questionnaire, Kuckelkorn et al. (2021) found differences between professional and amateur musicians with higher values in openness and neuroticism as well as lower values in agreeability and conscientiousness in professional musicians compared to the amateur musicians.

Our results appear comprehensible due to the different situation and requirements in which professional musicians find themselves compared to non-professional musicians and choir singers. For example, professional musicians must be more conscientious in order to meet professional standards. At the same time, being open to new experiences is particularly well suited to the profession of orchestral musician, especially as a high degree of flexibility is required in the music profession.

Female musicians in our study showed higher values in neuroticism and agreeableness than male musicians. This was found similarly in other studies (Herman and Clark, 2023).

In the evaluation of the PQM scales, the amateur choir singers stood out in comparison to professional and non-professional orchestra musicians with the highest self-efficacy before and during the performance. Amateur choir singers were best able to cope with MPA during the performance and had the fewest symptoms of MPA after the performance. Ryan and Andrews (2009) were also able to show that choir singers suffer from MPA, but that performances in instrumental ensembles triggered stronger MPA than choir performances.

Possibly due to less experience and playing skills than the professional orchestra musicians, the non-professional orchestra musicians showed the strongest symptoms of MPA before and during the performance. Interestingly, the professional orchestra musicians had the lowest self-efficacy at all three time points of the performance and remained at the same level of symptoms of MPA, especially after the performance. As self-efficacy can be influenced by high expectations and perfectionism, these lower values among professional orchestra musicians can be linked to this. From the perspective of musicians’ medicine, it could be important for professional orchestra musicians to consciously relax in the situation after the concert and improve their self-efficacy through positive self-assessment.

The results of our study show a positive correlation between self-efficacy and extraversion in relationship to MPA. These results were generally supportive of the theoretically expected relationships between personality traits and MPA in particular and are in accordance with Rodriguez-Mora and Lopez Diaz (2020).

Similarly, the findings on the correlation of higher levels of neuroticism and stronger MPA in other studies (Valentine et al., 1995; Smith and Rickard, 2004; Langendörfer et al., 2006; Thomas and Nettelbeck, 2014; Chattin, 2019) were confirmed. Additionally, musicians with higher neuroticism rated their performance lower. Musicians with higher values of extraversion rated their performance as better.

The relation of personality traits and aspects of MPA at a performance differed between amateur and professional musicians. Interestingly, among the professional musicians, neuroticism was only associated with the post-performance situation. This could indicate that personality traits play a particularly important role in the processing of performance experiences after the performance and would again focus in particular on the situation after the performance with professional orchestra musicians.

The musicians were divided into three MPA types according to Spahn et al. (2021) and in accordance with Spahn et al. (2023) and Papageorgi (2021). Musicians of Type 1 differed significantly from musicians of the other two Types 2 and/or 3 in the personality traits by higher extraversion, openness and conscientiousness and lower values for neuroticism.

In particular, Type 1 differed from the other two types in that openness and conscientiousness were stronger here. Chattin (2019) also found both personality traits to be able to reduce MPA. This could be interpreted to the effect that these musicians have more flexibility in the performance situation while at the same time being more conscientious in their preparation.

The correlations between low neuroticism and adaptive MPA from previous studies (Smith and Rickard, 2004; Chattin, 2019; Rodriguez-Mora and Lopez Diaz, 2020; Herman and Clark, 2023) are thus also confirmed with regard to a specific performance situation. The proximity between anxiety and neuroticism also makes it possible to relate the present results to the study results of Papageorgi (2021).

Even though the present study is based on a large sample, it should be borne in mind that the sample is rather heterogeneous, consisting of professional and non-professional musicians, as well as orchestra musicians and choir singers. The results obtained here should therefore be replicated in large, homogeneous samples.

Furthermore, the PQM was applied for each musician only once and for one performance. It might be interesting to investigate how personality traits may affect more than one performance of the same person.