Music Performance Anxiety

Aus dem Freiburger Institut für Musikermedizin Albert-Ludwigs-Universität Freiburg i. Br. (Prof. Dr. med. C. Spahn) Music Performance Anxiety A Review of the Literature INAUGURAL DISSERTATION zur Erlangung des Medizinischen Doktorgrades der Medizinischen Fakultät der Albert-Ludwigs-Universität Freiburg i. Br. Vorgelegt 2009 von Ariadna Ortiz Brugués geboren in Sevilla (Spanien)

Dekan: Prof. Dr. med. Christoph Peters 1. Gutachter: Prof. Dr. med. Claudia Spahn 2. Gutachter: Prof. Dr. med. Antje Aschendorff Jahr der Promotion: 2009

Index Introduction: Music Performance Anxiety definition of the term... Page 1 Literature Review... Page 7 I.A. Epidemiology...... Page 8 1. Children musicians... Page 8 2. Adolescent musicians... Page 13 3. Orchestra musicians... Page 20 4. Singers... Page 23 5. Popular musicians... Page 26 6. Music vs. non-music students... Page 27 7. Athletes vs. Musicians... Page 28 I.B. Methodical aspects... Page 30 I.C. Interventional studies... Page 33 1. Behavioral interventions... Page 33 2. Cognitive interventions... Page 39 3. Cognitive-behavioral interventions... Page 40 4. Combined interventions... Page 43 5. Other interventions... Page 51 6. Drug interventions... Page 61 7. Treatment reviews.. Page 75 I.D. Concept... Page 81 1. How Performance Anxiety occurs... Page 81 2. Coping strategies... Page 100 3. Psychology... Page 107 4. Predictors... Page 119 5. Context conditions... Page 132 Conclusions: 1. Suggestions for the correct use of the term MPA... Page 137 2. Identification of required studies... Page 137 3. Critic of the studies... Page 139 Summary... Page 142 Assorted reports... Page 143 Bibliography... Page 157 Appendix. Page 169

Nothing is more devastating to a performing artist than not having the chance to be on stage and, as the pervasiveness of performance anxiety attests, nothing is more threatening than having that chance (Plaut, 1990)

MUSIC PERFORMANCE ANXIETY Introduction: Music Performance Anxiety definition of the term Using the latest definition given by Kenny, Music Performance Anxiety (MPA) is the experience of marked and persistent anxious apprehension related to musical performance that has arisen through specific anxiety conditioning experiences and which is manifested through combinations of affective, cognitive, somatic and behavioral symptoms. It may occur in a range of performance settings, but is usually more severe in settings involving high ego investment and evaluative threat. It may be focal (i.e. focused only on music performance) or occur comorbidly with other anxiety disorders, in particular social phobia. It affects musicians their entire lives and is at least partially independent of years of training, practice and level of musical accomplishment. It may or may not impair the quality of the musical performance (Kenny, 2008). However, the definition of MPA has not always been so clear. It can be found in the literature that Performance Anxiety is often confused with the term Stage Fright, and some authors use both terms interchangeably (Salmon, 1990). Stage Fright affects 80% of the population when they are required to perform in front of an audience. It may affect musicians, singers, dancers, athletes, actors, public speakers, or students when test-taking. Thus, stage fright is, a normal reaction which should be optimised to enhance accomplishments in the performance situation. People affected by Stage Fright may report some of the following symptoms: psychological (doubts about one s self, failure expectancy, or catastrophizing), physical (such as higher breath and heart frequency, dry mouth, shaking, sweating palms, etc., as a consequence of sympathetic nervous system hyperactivation) and behavioral (a combination of activeness and motivation on the one hand, and retraction and avoidance on the other hand). Depending on how intense these symptoms are and how they interfere with the performance, we talk about Performance Anxiety. In this case, those aspects of Stage Fright which diminish accomplishment outweigh the positive facets, thus necessitating treatment (Spahn, 2006). Consequently, it is essential that authors use both terms adequately. While Stage Fright refers to a normal reaction, Performance Anxiety is a pathological disorder and necessitates treatment. 1

The definition of MPA by Kenny is based on Barlow s (2000) model of anxiety, an integrated set of triple vulnerabilities: generalized biological (heritable), generalized psychological (early experiences in developing a sense of control over salient events), and more specific psychological vulnerabilities (anxiety associated with certain environmental stimuli through learning processes). According to this model, the systems affected in anxiety are the somatic (hyperarousal or acute stress response that produces a range of bodily sensations that prepare the body to meet the perceived challenge), the emotional (anxiety, fear, panic), the cognitive (worry, dread, inattention and distractibility, lack of concentration, memory loss), and the behavioural manifestations (technical errors, memory loss, performance breaks, avoidance of performance opportunities) (Kenny, 2008). Performance anxiety usually occurs as an isolated disorder, affecting only one specific part of a person s life (Kenny & Ackermann, 2007). However, for a significant minority, other co-morbid disorders may be present. The most common of them is generalised anxiety disorder, which appears to co-occur in about one third of those presenting with severe performance anxiety (Sanderson, DiNardo, Rapee & Barlow, 1990). Such individuals have a long history of generalised worry and apprehension in most facets of their lives, not solely in situations requiring performance. Others may qualify for a diagnosis of social phobia (social anxiety) if they demonstrate significant impairment in interactions with others as well as the performance setting, while meeting the criteria for social phobia presented in DSM IV (APA 1994). About 10-15% of those also meet criteria for clinical depression (Kessler, Stang, Wittchen, Stein & Walters, 1999). For a small group of sufferers, there may be underlying psychological conflicts. One of the more common of these is the "imposter syndrome (Lazarus & Abramovitz, 2004) whereby the individual fears that an individual will eventually be exposed as a fraud, with every performance potentially confirming his underlying fear. However, the highly anxious individual does not necessarily demonstrate impaired or inferior performance compared to low anxious individuals (Strahan & Conger, 1998). Anxiety is a two-factor structure, with both state and trait components. State anxiety is a transitory emotional state characterised by heightened tension and apprehension. Trait anxiety refers to relatively stable individual differences between people in their tendencies to respond to situations perceived as threatening with 2

elevations in state anxiety. Anxiety occurs on a continuum from mild to severe. It has also different forms: adaptative, reactive, maladaptative, and pathological. In adaptative anxiety, the body adapts to a threatening or challenging situation by increasing the state of arousal. This type of anxiety may be experienced as excitement and may enhance coping and improve performance. Reactive anxiety results from actual or perceived inability to meet the demands of the situation. Maladaptative anxiety impairs thinking and problemsolving and has a negative effect on behaviour or performance. Pathological anxiety occurs in situations in which the individual can not identify the cause of the anxiety. This state is often referred to as Generalised Anxiety Disorder (Kenny, 2008). Several theories have been proposed to explain the origins of anxiety. A family environment characterised by limited opportunity for personal control is associated with the development of anxiety (Chorpita, Brown, & Barlow, 1998). Building a positive learning history during childhood by providing opportunities to cope adaptatively with challenges should immunise children against the development of anxiety as a response to subsequent negative learning episodes (Barlow, 2002; Field, 2006). Ehlers (2003) showed that people with panic attacks report observing more panic behaviours in their parents than people with other anxiety disorders and people with no anxiety disorder. Many people who develop anxiety disorders report early-learning experiences as children (Chambless, Caputo, Gright, & Gallagher, 1984). Recent research has identified various roles for different neural substrates in fear conditioning (Kenny, 2008): the amygdala is involved in the physiological response to conditioned fear (Gazzaniga, Irvy, & Magnun, 2002), while the hippocampus is involved in the associative learning component (Squire & Zola-Morgan, 1991). Very young children rarely experience performance anxiety; on the contrary, they usually love to perform. Therefore, a transition takes place between childhood (not presenting MPA) and adolescence (specifically from 14 to 19 years old; Osborne, Kenny, & Holsomback, 2005) and adults suffering from MPA. This transition is due to a combination of factors: innate temperament, increasing cognitive capacity and self-reflective function, type of parenting and other interpersonal experiences, perception and interpretation of surroundings, technical skill and mastery, and specific performance experiences that may have positive or negative outcomes. Furthermore, it has been proved that the capacity for 3

self-evaluation (self-criticism) emerges in middle to late adolescence (Jackson & Lurie, 2006). Some authors classify performance anxiety as a subcategory of social phobia (Hook & Valentine, 2002; Turner, Johnson, Beidel, Heiser, & Lydiard, 2003). According to the American Psychiatric Association, social phobia is defined as a marked and persistent fear of one or more types of social or performance situations in which the person is exposed to unfamiliar people or to possible scrutiny by others. The individual fears that he/she will act in a way (or show anxiety symptoms) that will be humiliating or embarrassing (DSM-IV-TR, 300.23) (APA, 2000). The classification of social phobia into generalized (anxiety is experienced about interpersonal interactions generally), nongeneralized (anxiety is experienced in settings in which the individual is being scrutinized), and specific (anxiety occurs for very few performance situations) sub-types (Turner, Johnson, Beidel, Heiser, & Lydiard, 2003) may help in the clarification of MPA as a form of specific social phobia (Kenny, 2008). However, several differences exist between social phobia and performance anxiety. People suffering from performance anxiety are more likely to have higher expectations of themselves (Abbot & Rapee, 2004), or a greater fear of their own evaluation of their performance. Social phobia, on the other hand, is fear of scrutiny from others (Stoebert & Eismann, 2007), though it is is also present in MPA as well as a higher degree of post-event rumination (Abbott & Rapee, 2004) and a continued commitment to the feared performance situation, as opposed to avoidance of the feared situation in social phobia (Powell, 2004). Furthermore, in social phobia the audience is often imaginary. In other words, socially phobic individuals fear that everyone is watching and judging them, when the reality may be that the person has not been noticed in the feared social setting (Kenny, 2008). For the artistic or sports performer, the audience is real and performers are usually correct in their assessment that people are watching and judging them (Brotons, 1994). Finally, aspiring and professional musicians are highly invested in their identities as musicians, and find it difficult to disentangle their selfesteem from their musical self-efficacy (Kemp, 1996). This fact makes musicians and other high level performers more vulnerable to anxiety because of the perception that failure as a performing artist signifies failure as an individual. (Chesky & Hipple, 1997). 4

An optimal performance is determined by a complex interaction between person characteristics, task characterisics, and performance setting (Kenny, 2008). Person characteristics are mainly determined by anxiety (understood as the state experienced when one believes that the demands are excessive or unachievable) and perfectionism. Perfectionism has both positive and negative sides (Stoebert & Eismann, 2007). Striving for perfection is associated with intrinsic motivation, higher effort, and higher achievement, whereas perceived pressure is associated with intrinsic motivation, extrinsic motivation and higher distress. Task characteristics contain task complexity and task mastery (Kenny & Ackermann, 2007). Task complexity can be explained from the Yerkes- Dodson Inverted U Curve (Yerkes & Dodson, 1908), which shows the relationship between the amount of physiological arousal experienced by a performer and the quality of his/her performance. Optimal performance on simple tasks will increase as arousal increases, but will deteriorate on complex tasks after a moderate level of arousal is exceeded. With task mastery, a similar relationship has been observed between the degree to which the task has been practiced to achieve mastery (automaticity) and the amount of physiological arousal needed to produce an optimal performance (Kenny, 2008). Even at ideal levels of arousal, low practice will produce a suboptimal performance. A well-practised piece will be performed well at higher levels of arousal but will also suffer if arousal exceeds a certain optimal maximum (Kokotsaki & Davidson, 2003). Focused attention shows a similar relationship, increasing when there are low to moderate levels of arousal but decreasing with very high levels of arousal (Kenny & Ackermann, 2007). Finally, performance setting influences the level of performance. Musicians are more likely to feel anxious under conditions of evaluation, jury performances and concert/recitals than they would under practice conditions (Kenny & Ackermann, 2007). Most forms of performance anxiety are difficult to treat, and anxiety levels after treatment rarely reduce to those of non-anxious people (Kenny & Ackermann, 2007). The best form of treatment is to prevent its occurence. Awareness of the availability of effective treatments for musicians with performance anxiety should be introduced to student musicians at an early stage of their musical training. Sound pedagogy, appropriate parental support and expectations, and the learning of self-management strategies early in one s musical education can help to mitigate the effects of entering a highly stressful profession. Repeated exposure to the feared situation (music performance) in the absence 5

of the development of skills and strategies to ensure success is likely to have a detrimental effect on the performer with potentially devastating consequences (Kenny, 2008). Repertoire should be well within the technical capacity and interpretative abilities of the student and the material should be over-learned to the point of automaticity (Kenny & Ackermann, 2007). In order to have a better understanding of Musical Performance Anxiety, a review of the literature has been done and it is presented below. Music Performance Anxiety has been proven to arise independently of age, experience and performance setting (Kenny, 2006). Great musicians like Pau Casals (Plaut, 1990) or Enrico Caruso (Spahn, 2006) suffered from MPA. Females are more affected than males (Nagel, 1988; LeBlanc et al., 1997; Miller et al., 2004; Rae et al., 2004; and Osborne et al., 2008). Solo performances showed higher MPA scores than ensemble performances (Simon et al., 1979; Rife et al., 2000). Jury evaluations produce higher levels of MPA, compared to non-evaluation context situations (Craske et al., 1984; Salmon et al., 1989; Rae et al., 2004; and Yoshie et al., 2008). Catastrophizing has been found to be the best predictor of MPA (Zinn et al., 2000; Liston et al., 2003) and an association between arm stiffness and MPA has recently been reported (Yoshie et al., 2008). Both facts may contribute to the reason why Cognitive- Behavioral Interventions have proved to have good results on treating MPA, as they focus on changing faulty thinking patterns that give rise to maladaptive behaviors, as well as changing the dysfunctional behaviors that arise when people feel anxious (being excessive muscle tension the main symptom) (Kenny, 2004). This review classifies studies under four main elements (Epidemiology, Methods, Intervention, and Concept) and analised with the Evidence-based medicine (EBM) criteria. Differences in MPA will be shown among age groups, gender, and solo vs. ensemble performances. Scales that measure MPA are also provided. A review of the treatments for MPA was done, also, focusing on those that have proven to be more effective. A number of coping strategies, predictors, and varying context situations, among other interesting facts, are also analised. At the end of the review, conclusions regarding how studies were performed are provided. 6

Literature Review Evidence-based medicine (EBM) aims to apply evidence gained from the scientific method to certain parts of medical practice. It seeks to assess the quality of evidence relevant to the risks and benefits of treatments (including lack of treatment). According to the Centre for Evidence-Based Medicine, "Evidence-based medicine is the conscientious, explicit and judicious use of current best evidence in making decisions about the care of individual patients." EBM recognizes that many aspects of medical care depend on individual factors such as quality and value-of-life judgments, which are only partially subject to scientific methods. EBM, however, seeks to clarify those parts of medical practice that are in principle subject to scientific methods and to apply these methods to ensure the best prediction of outcomes in medical treatment, even as debate about which outcomes are desirable continues. The foundation of evidence-based medicine is the systematic review of evidence for particular treatments, mainly randomized controlled trials. The Cochrane Collaboration leads this effort. A 2001 review of 160 Cochrane systematic reviews in the 1998 database revealed that, according to two readers, 41.3% concluded positive or possibly positive effect, 20% concluded evidence of no effect, 8.1% concluded net harmful effects, and 21.3% of the reviews concluded insufficient evidence. A review of 145 alternative medicine Cochrane reviews using the more up-to-date 2004 database revealed that 38.4% concluded positive effect or possibly positive (12.4%) effect, 4.8% concluded no effect, 0.69% concluded harmful effect, and 56.6% concluded insufficient evidence. Evidence-based medicine categorizes different types of clinical evidence and ranks them according to the strength of their freedom from the various biases that beset medical research. For example, the strongest evidence for therapeutic interventions is provided by systematic review of randomized, double-blind, placebo-controlled trials involving a homogeneous patient population and medical condition. In contrast, patient testimonials, case reports, and even expert opinion have little value as proof because of the placebo effect, the biases inherent in observation and reporting of cases, difficulties in ascertaining who is an expert, and more. Systems to stratify evidence by quality have been developed. In this review we are using the classification given by Richter, B., Zander, M., & Spahn, C. (2007): 7

Level I: Systematic review of randomized, double-blind, placebo-controlled trials. A meta-analysis combines the results of several studies that address a set of related research hypotheses. Level II: Evidence obtained from at least one properly designed randomized controlled trial. Level III: Evidence obtained from well-designed controlled trials without randomization. Level IV: Evidence obtained from non-experimental, non-randomized trials. Level V: Opinions of respected authorities, based on clinical experience, descriptive studies, or reports of expert committees. I.A. Epidemiology: 1. Children musicians: STUDIES EVALUATION CRITERIA Simon, J. A., & Martens, R. (1979): Children s anxiety in sport and nonsport evaluative activities. LeBlanc, A., Jin, Y. C., Obert, M., & Siivola, C. (1997): Effect of audience on MPA. Ryan, C. (1998): Exploring MPA in children. Ryan, C. (2004): Gender differences in children s experience of MPA. Ryan, C. (2005): Experience of MPA in elementary school children. Level III Level III Level III Level III Level III Evidence obtained from welldesigned controlled trials without randomization. Evidence obtained from welldesigned controlled trials without randomization. Evidence obtained from welldesigned controlled trials without randomization. Evidence obtained from welldesigned controlled trials without randomization. Evidence obtained from welldesigned controlled trials without randomization. 8

Author Year Subject type Methods Conclusions Publication Study design Simon, J. 1979 749 9-14 year old Quantitative: The greatest anxiety A., & Journal of boys, comparing Competitive was reported by boys Martens, Sport anxiety in test, sport State performing solo on a R. Psychology, and musical activities. Anxiety musical instrument; 1, 160-169 Non-randomized Inventory performing with a study. (CSAI), a 10- band was responsible item for the highest shortened anxiety among group version of activities. Spielberger s (1973) State Anxiety Inventory for Children. LeBlanc, 1997 27 male and female Quantitative: MPA increases with A., Jin, Y. Journal of high school band an analog audience size and C., Obert, Research in members performing scale self- perceived importance M., & Music solos under 3 levels of report of of the performance. Siivola, C. Education, audience presence: performance Heart rate was 45, 480-496 alone in a practice anxiety, steady across the room, in a practice heart rate first two room with one recorded performance researcher present, during conditions, but rose and in the rehearsal performance, significantly at the room with all judges third. Females researchers, a peer rating of the presented better group, and a tape final performances, but recording being made. performance. reported significantly Non-randomized Qualitative: higher anxiety levels 9

study. an exit than males in the interview. first and third performance conditions. Ryan, C. 1998 26 12-year-old piano Quantitative Similar physical and Medical students, by (heart rate physiological Problems of comparing heart rates monitoring, symptoms of MPA as Performing monitored STAIC, and adult musicians. Artists, continuously through a Coopersmith Significant increases 13(3), 83-88 piano lesson and a Self-Esteem in heart rate between recital performance Inventory) baseline and recital, for each subject. The and and between the state portion of the Qualitative "sitting stage-side," State-Trait Anxiety (interviews). "walking onstage," and Inventory for "playing" stages of Children (STAIC) was the recital. administered to Significant subjects at the correlations between recital. The trait self-esteem, portion of the STAIC, particularly social the Coopersmith Self- self-esteem, and both Esteem Inventory, state anxiety and and individual trait anxiety. interviews were Seventeen of the 22 completed during non- subjects interviewed performance times reported feelings of before and after the anxiety when recital. performing in piano Non-randomized recitals, many noting study. a fear of making mistakes in front of people as the primary cause. 10

Ryan, C. 2004 26 grade students Quantitative Girls had substantially Psychology under 12 years old (heart higher heart rates of Music, 32, performing in a piano monitoring, than boys immediately 89-103 recital were STAI-C) and prior to but not monitored Qualitative during their continuously on (interviews). performance; boys measures of heart had significantly more rate and behaviour. anxious behaviours They were than girls both prior interviewed in the to, and during a months prior to the performance. recital and they completed the State- Trait Anxiety Inventory for Children (STAI-C) immediately after performing. Non-randomized study. Ryan, C. 2005 173 3-7 year old Quantitative State anxiety was International children who (trait and significantly higher on Journal of completed the trait state form the day of the school Stress and state form of the of STAIC). concert and was Management STAIC during a related to children s regular school day and level of trait anxiety. the state form again on the day of a major school concert. 11

Simon et al. compared the levels of anxiety among children between 9 and 14 years old under different performance tasks (test, sport, and music). They found that music performance was responsible for the highest levels of anxiety. LeBlanc et al. showed that MPA and heart rate increases with audience size and perceived importance of the performance. They also focused on possible gender differences, finding that females presented better performances, but reported significantly higher anxiety levels than males. Ryan (1998) used a group of 12-year-old piano students, finding that they presented similar physical and physiological symptoms of MPA as adult musicians. Many of them noted a fear of making mistakes in front of people as the primary cause. This study also provided significant correlations between self-esteem, particularly social self-esteem, and both state anxiety and trait anxiety. The results from heart rate studies agreed with those shown by LeBlanc et al. Focusing on gender differences, Ryan (2004) found that girls had higher heart rates than boys immediately prior to but not during their performances, while boys had more anxious behaviours than girls both prior to and during a performance. This finding conflicts with LeBlanc s results that females had higher anxiety levels than males. However, the group ages vary between both studies, so it could be concluded that boys at the age of 12 present more anxious behaviours than their female peers, both prior to and during a performance. High school females, on the contrary, have higher performance anxiety levels than high school males. In a sample of 3 to 7 year-old children, Ryan (2005) found that state anxiety was significantly higher on the day of the school concert than during a regular school day and that this state anxiety was related to a child s level of trait anxiety. This study shows, surprisingly, the existence of performance anxiety among children between the ages of 3 and 7. The studies presented here are well-designed and show very interesting results. The number of subjects evaluated is consistent, and the group ages are clearly defined. None of them is randomized, however, so additional well-designed randomized studies would be beneficial, where a well-designed study has large enough samples, clear group age designs, and results showing possible gender differences. 12

2. Adolescent musicians: STUDIES EVALUATION CRITERIA Shoup, D. (1995): Survey of performance-related problems among high school and junior high school musicians. Britsch, L. (2005): Investigating performance-related problems of young musicians. Fehm, L., & Schmidt, K. (2006): Performance anxiety in gifted adolescent musicians. Kenny, D. T., & Osborne, M. S. (2006): MPA - new insights from young musicians. Osborne, M. S., & Kenny, D. T. (2006): Impact of a music performance enhancement program on MPA in secondary school music students. Osborne, M. S. & Kenny, D. T. (2008): The role of sensitizing experiences in MPA in adolescent musicians. Level IV Level IV Level IV Level III Level II Level III Evidence obtained from nonexperimental, non-randomized trials. Evidence obtained from nonexperimental, non-randomized trials. Evidence obtained from nonexperimental, non-randomized trials. Evidence obtained from welldesigned controlled trials without randomization. Evidence obtained from at least one properly designed randomized controlled trial. Evidence obtained from welldesigned controlled trials without randomization. Author Year Subject type Methods Conclusions Publication Study design Shoup, D. 1995 425 high Quantitative 33.2% indicated having a Medical school and (survey). MS performance-related Problems of junior high problem at some point Performing school band since they began playing. Artists, 10(3), and orchestra Of these students, 51.9% 100-105 students were reported the problem as 13

surveyed about performancerelated problems. Nonrandomized study. current, yielding a prevalence of 19.6%. However, only 7.5% had missed rehearsals and only 12.5% had had to refrain from playing for more than one week. Consequently, most students reported a low severity grade for their problems. The most common treatment used was rest. 44.0% believed that they should continue playing while in pain ( no pain, no gain ). About performance anxiety, 55.5% indicated having symptoms that negatively affected their performance (approximately the same proportion as in professional musicians). More than 18% admitted to severe nervousness with significant impact on their performance. 48.9% of the students complained of one or more non-ms problems that negatively affected their playing. The most 14

common were asthma, eye strain, headaches, cold sores, and allergies. Britsch, L. 2005 97 students in Qualitative Few reported accepting Medical four youth (interviews). the belief of playing Problems of orchestras in a through pain, while Performing midsized performance anxiety was Artists, 20(1), midwestern reported at a higher- 40-47 city (USA) than-expected rate. The were surveyed. older students had more Non- awareness of possible randomized causes of playing-related study. pain and strategies for its elimination than did the younger students. The older students also discussed their pain with teachers more frequently, receiving information that was beneficial in reducing or solving the problem. Two statistically significant relationships were found: one between amount of practice time and grade of pain reported in the youngest orchestra, and the other comparing females by age and grade of pain reported. Fehm, L., & 2006 74 15-19-year- Quantitative About one third of the 15

Schmidt, K. Journal of old pupils who (survey). group were distinctly anxiety attended a handicapped by their disorders, German special performance anxiety. 20(1), 98-109 music school. Unfavourable coping Non- strategies, such as drug randomized of alcohol abuse were study. rarely reported. Most pupils called for more support either from their teachers of from outside of school to cope with their anxiety. Kenny, D. T. 2006 381 young Quantitative MPA was more & Osborne, Advances in musicians aged (MPAI-A). specifically related to M. S. Cognitive 12 to 19 years social anxiety than trait Psychology, attending anxiety with stronger 2(2-3), 103- secondary high positive correlations 112 schools. between the MPAI-A and Somatic and social phobia measures Cognitive than MPAI-A and trait Features, anxiety. The cognitive Performance component of anxiety Context, and contributed to the Performance prediction of music Evaluation performance anxiety were more than the somatic investigated. component. Non- The experience of MPA randomized may begin early in a study. musical career and the characteristics of this experience are qualitatively similar to 16

those experienced by adult musicians. Osborne, M. 2006 23 adolescents Quantitative: Significant improvements S. & Kenny, Musicae with high MPA self-reports in self-reported MPA D. T. Scientiae, 1-17 from a of MPA, trait were observed at post- selective high and state test for adherent school were anxiety, students only (i.e., randomly diagnostic students who were assigned to interview for actively engaged in the either a social phobia, program and who adopted seven-session heart rate, program techniques). intervention frontalis EMG, Adherent students also program or a and had higher MPA at behavior- performance commencement. Non- exposure-only quality. adherent and behavior- control group. exposure-only students The both showed reductions intervention in MPA over the study consisted of period but not to the psycho- same degree as adherent education, goal students. There appeared setting, to be no effect of CBT on cognitive performance quality. restructuring, relaxation training and behavioral exposure in the form of two solo performances with audience. Randomized 17

study. Osborne, M. 2008 298 music Quantitative: Both hypotheses were S. & Kenny, Psychology of students were Descriptions supported: first, that D. T. Music, 1-16 asked to were scored negative cognitions were provide according to more predictive of written six domains - adolescent MPA than the descriptions situational and behavioral, somatic or of their worst behavioral affective components of performance, factors, the sensitizing what happened affective, experience; second, that and how they cognitive and music students who felt, somatic reported a negative music specifying symptoms of performance experience their age at anxiety, and self-reported higher that time, outcome. levels of MPA than those audience Demographics, who did not report such members, and MPAI-A, and an experience. any events STAI-T were MPA was best predicted that occurred also used. by trait anxiety and subsequent to gender. Females reported the more emotional distress performance. than males and had Non- significantly higher total randomized scores. study. The survey conducted by Shoup among a respectable number of high school music students showed that the proportion of students indicating symptoms of performance anxiety was approximately the same as in professional musicians, with more than 18% of the students admitting these symptoms to be severe enough to have a negative impact on their performance. The proportion of students that indicated having a musculoskeletal performance-related problem was lower than that of professional musicians (Fishbein et 18

al., 1988; Raeburn et al., 2003). This may be due to the fact that professional musicians have usually spent more years practicing than young musicians and, consequently, their muscles are more affected. Alarmingly, approximately half of the student musicians who reported a MS problem believed that they should keep playing while in pain. Fortunately, the study conducted by Britsch among musicians playing at youth orchestras showed that few of them reported accepting the belief of playing through pain. This fact can be explained by greater maturity from age and experience, as Britsch also found that the older students were more aware of possible causes of playing-related pain and strategies for its elimination than the younger students. They also admitted discussing their pain with teachers more frequently, receiving information that was beneficial in reducing or solving the problem. Performance anxiety was also reported frequently. There was a relationship between the amount of practice time and the grade of pain reported, which supports the idea mentioned above that musculoskeletal problems are related to years of practice. The survey taken by Fehm et al. among a group of 15-19 year-old music students reported that about one third were handicapped by their performance anxiety, most of them calling for more support from their private teachers to cope with anxiety. Kenny et al. studied a large group of music students between 15 and 19 years old, and found that MPA was more closely related to social anxiety than trait anxiety. They discovered that the cognitive component of anxiety contributed to the prediction of music performance anxiety more than the somatic component. They concluded that the experience of MPA may begin early in a musical career and the characteristics of this experience are qualitatively similar to those experienced by adult musicians. This last statement matches the study conducted by Ryan (1998) with children musicians. Osborne et al. (2006) selected a group of adolescent musicians who were randomly assigned to either an intervention program (psycho-education, goal setting, cognitive restructuring, relaxation training and behavioral exposure in the form of two solo performances with audience) or a control group. Those students who were assigned to the intervention program showed significant improvements in self-reported MPA. However, there appeared to be no effect of Cognitive Behavioral Therapy (CBT) on performance quality. Osborne et al. (2008), from written descriptions given by a good sample of music students, concluded that negative cognitions were more predictive of adolescent MPA than the behavioral, somatic or affective components of the sensitizing experience. They also concluded that music students who reported a negative music performance experience self-reported higher 19

levels of MPA than those who did not report such an experience. MPA was found to be best predicted by trait anxiety and gender, with females reporting more emotional distress than males and having significantly higher total scores. This fact matches with the study by LeBlanc et al., who also found that females (high school music students) had higher anxiety levels than males under performance conditions. Although the results of these studies are promising, only the one by Osborne et al. (2006) is randomized, so it would be desirable to have more randomized studies with adolescent musicians of different ages and gender. 3. Orchestra musicians: STUDIES EVALUATION CRITERIA Fishbein, M., Middlestadt, S. E., Ottati, V., Straus, S., & Ellis, A. (1988): Medical problems among ICSOM musicians overview of a national survey. Lockwood, A. H. (1989): Medical problems of musicians. van Kemenade, J. F., van Son, M. J., & van Heesch, N. C. (1995): Performance anxiety among professional musicians in symphonic orchestras a self-report study. James, I. (1998): Western orchestral musicians are highly stressed. Level IV Level IV Level IV Level IV Evidence obtained from nonexperimental, non-randomized trials. Evidence obtained from nonexperimental, non-randomized trials. Evidence obtained from nonexperimental, non-randomized trials. Evidence obtained from nonexperimental, non-randomized trials. 20

Author Year Subject type Methods Conclusions Publication Study design Fishbein, M., 1988 National Quantitative 82% experienced Middlestadt, Medical survey of (survey). medical problems, 76% S. E., Problems of 2122 stated that at least one Ottati, V., Performing professional medical condition Straus, S., & Artists, 3(1), orchestral interfered with their Ellis, A. 1-8 musicians at performances, and 36% the reported suffering from International up to four independent Conference of problems. Symphony and Most prevalent problems Opera reported: lower back Musicians (22%), neck (22%), (ICSOM). shoulder (20%), and Non- upper back (16%). Stage randomized fright (24%), depression study. (17%), sleep disturbances (14%), acute anxiety (13%), and severe headaches (10%). Lockwood, 1989 Survey of 48 Quantitative 24% of musicians A. H. New England orchestras (survey). frequently suffered Journal of (International stage fright (defined in Medicine, Conference of this study as the most 320, 221- Symphony and severe form of MPA), 227 Opera 13% experienced acute Musicians anxiety and 17% National US depression. survey). Nonrandomized study. 21

van 1995 155 of 650 Quantitative 91 of the 155 (58%) Kemenade, Psychological professional (survey). reported experiencing J. F., van Reports, 77, musicians or having experienced Son, M. J., & 555-562 playing performance anxiety van Heesch, symphonic seriously enough to N. C. orchestras in affect their The professional or personal Netherlands lives. There appeared to completed a be no difference in self-report prevalence between men questionnaire and women. Substantial concerning percentages of the performance anxious musicians anxiety. reported considerable Non- anticipation anxiety days randomized (36%), weeks (10%), or study. even months (5%) prior to a performance. James, I. 1998 Survey of 56 Quantitative 70% of musicians Resonance: orchestras. (survey). reported that they International Non- experienced anxiety Music randomized severe enough to Council, 26, study. interfere with their 19-20 performance, with 16% experiencing this level of anxiety more than once a week. All of the studies here reviewed are level IV, so it would be desirable to take a randomised trial with a good sample of orchestra musicians. Four studies have an appropriate number of participants, but they are all based solely on surveys and none of them is recent. Both Fishbein et al. and Lockwood studies are based on the survey from 22

the International Conference of Symphony and Opera Musicians, reporting 24% of cases suffering from stage fright. The number increases to 58% in the study by van Kemenade et al., and 70% in the one by James. This can be attributed to differing terms used in each study. Fishbein et al. and Lockwood use the term stage fright, while van Kemenade et al. write about performance anxiety, and James quotes having experienced anxiety severe enough to interfere with their performance. Furthermore, there is a difference of ten years time between the first study (Fishbein et al., 1988) and the last one (James, 1998). It is well-known that musicians are typically ashamed of admitting to be suffering from performance anxiety. Moreover, in 1988 very few studies focused on therapies as a means to deal with performance anxiety, while in 1998 more studies about the subject had been written. The more that musicians are educated about performance anxiety, the more readily they can recognize its prevalence in other musicians without feeling ashamed. Along with performance anxiety, Fishbein et al. reported muscle pain among musicians, lower-back and neck being the most affected parts of the body. Among other psychical disturbances, depression and acute anxiety were highly reported. Interestingly, van Kemenade et al. showed that there was no difference between men and women reporting performance anxiety. Moreover, they found a substantial percentage of musicians who displayed symptoms of anxiety for days, weeks, and even months prior to a performance. This matches the survey taken by James, who found 16% of the sample having the same level of anxiety that they experience during a performance more than once a week. 4. Singers: STUDIES EVALUATION CRITERIA Sandgren, M. (2002): Voice, Soma, and Psyche. Kenny, D. T., Davis, P., & Oates, J. (2004): MPA and occupational stress Level IV Level III Evidence obtained from nonexperimental, non-randomized trials. Evidence obtained from welldesigned controlled trials 23

amongst opera chorus artists and their without randomization. relationship with state and trait anxiety and perfectionism. Author Year Subject type Methods Conclusions Publication Study design Sandgren, 2002 15 opera Qualitative Psychological problems were M. Medical singers were (interviews) most frequent: preoccupation Problems asked to and with the risk of vocal of report quantitative: indisposition, the habit of Performing problems the results of testing the voice; physical Artists, related to the interviews problems centered on 17(1), 11- their were used to respiratory tract symptoms; 21 professional construct a psychosocial problems work, coping self- concerned most difficulties in strategies, administered maintaining a familiy life due motivational 138-item to travel and irregular factors, and questionnaire working hours. Significant strongly with two open- positive correlations were emotional ended found between worry about singing questions. other s opinions and a number experiences. of variables: somatic Non- problems, testing the voice, randomized depression, and performance study. anxiety. Kenny, D. 2004 Relationships Quantitative: Scores indicating high trait T., Davis, Journal of among state Spielberger s anxiety (from STAI) were P., & Anxiety and trait State Trait approximately three times Oates, J. Disorders, anxiety, Anxiety (50%) more prevalent among 18, 78-82 occupational Inventory opera chorus artists than stress, (STAI), Cox among the normative sample perfectionism, and Kenardy for the test (15%). Anxiety 24

aspiration, and music performance anxiety were studied in a group of elite operatic chorus artists (32 participants) employed fulltime by a national opera company. Nonrandomized study. MPA Scale (CK-MPA) (modified to include separate scales for solo and choral performance), Occupational Stress Inventory- Revised (OSI- R), Kenny Music Performance Anxiety Inventory (K- MPAI), and Frost Perfectionism Scale (FROST-PE). was not related to occupational roles or issues related to the physical environment or working conditions. These results suggest that occupational stress makes a separate contribution to the quality of working life experienced by elite choral artists. Both studies showed that singers, as well as instrumentalists, suffer from performance anxiety, depression, and other somatic and psychosocial problems, as reported by Sandgren. Physical disorders differ from other musicians; due to the nature of their instrument since they primarily use the respiratory tract. The study from Kenny et al. applies several scales for performance anxiety that are also used with instrumentalists which prove performance anxiety among singers, making a connection between performance anxiety and other elements like trait anxiety, occupational stress, and perfectionism. Larger samples as well as randomised studies are desirable. 25

5. Popular musicians: STUDIES EVALUATION CRITERIA Raeburn, S. D., Hipple, J., Delaney, W., & Chesky, K. (2003): Surveying Popular Musicians health status using convenience samples. Level IV (Sample 1) Level II (Sample 2) Evidence obtained from nonexperimental, nonrandomized trials (Sample 1). Evidence obtained from at least one properly designed randomized controlled trial (Sample 2). Author Year Subject type Methods Conclusions Publication Study design Raeburn, 2003 Sample 1: 111 Quantitative Sample 1: 26% cited a S. D., Medical popular (non- current medical problem, 49% Hipple, J., Problems musicians standardized had at least one non- Delaney, of attending one questionnaires musculoskeletal (N-MS) W., & Performing of the three developped by problem, 74% had at least Chesky, K. Artists, regional music the authors). one MS problem, and 42% 18(3), 113- conferences reported hearing loss. 119 (Portland, Sample 2: 37% reported Austin or San hearing loss. Francisco). Depression and anxiety were Non- among the most frequently randomized cited N-MS problems for study. both samples. Pain and Sample 2: 115 stiffness were the most popular frequently reported MS musicians from problems for both samples. 26

a random sample of a musicians union list. Randomized study. 10% of Sample 1 musicians and 16% of Sample 2 musicians indicated that alcohol or drug use had hurt their performance in the previous year. Interestingly, the authors found depression and anxiety to be the most frequent non-musculoskeletal problem cited by popular musicians, which agrees with the results from both Fishbein et al. and Lockwood surveys. This also coincides with the fact that pain and stiffness appear to be the most frequently reported musculoskeletal problems. The present study also showed alcohol or drug abuse among 16% of the sample. This statistic is also known among some classical musicians who continue to take these substances to cope with performance anxiety. The use of standard scales in this study would be more reliable, in order to compare the results with other studies that used identical scales. 6. Music vs. non-music students: STUDIES EVALUATION CRITERIA Chesky, K. S. & Hipple, J. (1997): Performance anxiety, alcohol-related problems, and social/emotional difficulties of college students. Level IV Evidence obtained from nonexperimental, nonrandomized trials. Author Year Subject type Methods Conclusions Publication Study design Chesky, 1997 359 lower- Quantitative Analysis indicated significant K. S. & Medical division music (PAI, YAAPST, differences between the Hipple, Problems and non-music and a non- music and non-music major 27

J. of major students. standardized groups. The subjects Performing Measures social/emotional majoring in music were Artists, included two problem determined to have 12(4), 126- versions of the checklist test). significantly fewer alcohol- 132 PAI, the Young related problems and social- Adult Alcohol emotional concerns compared Problems with the non-music major Screening Test students. (YAAPST), and a social/emotional problem checklist. Nonrandomized study. This study shows the positive effects of music among adolescents. Those who study music showed having fewer alcohol- and social- related problems than those not studying music. The study used standardized and non-standardized scales, and the number of participants was very appropriate. 7. Athletes vs. musicians: STUDIES EVALUATION CRITERIA Lacaille, N., Whipple, N., & Koestner, R. (2005): Reevaluating the benefits of performance goals (musicians and athletes). Level IV Evidence obtained from nonexperimental, nonrandomized trials. 28

Author Year Subject type Methods Conclusions Publication Study design Lacaille, 2005 86 high-level Quantitative: Athletes: mastery- (in which N., Medical musicians and the the focus is on the Whipple, Problems 112 high-level Performance- development of skills) and N., & of athletes approach Goal performance- (in which the Koestner, Performing reported their Items, the focus is on demonstrating R. Artists, goals prior to a Performance- competence relative to 20(1), 11- peak avoidance Goal others) approach goals were 16 performance Items, the associated with optimal and a Mastery Goal functioning. Musicians: catastrophic Items, and the performance goals were performance. Intrinsic Goal markedly more detrimental Non- Items were than mastery goals. It was randomized used. also found that intrinsic goals study. associated with a focus on enjoyement were particularly helpful for musicians. It is interesting to compare how athletes and musicians react to pressures of their different professional lives. The present study reported that while athletes use mastery and performance goals in a positive way, performance goals are more detrimental than mastery goals among musicians. Nevertheless, intrinsic goals were found helpful for musicians. The results show that, although it is known that some athletes also suffer from performance anxiety, their goals prior to a performance vary substantially from those of musicians. Therefore, it can be concluded that therapies may also be substantially different for each group. 29