. 2003 by the American Music Therapy Association

Darcy DeLoach Walworth, MM, MT-BC ，

The Florida State University and Tallahassee Memorial Health Care

       The purpose of this study was to investigate the differences of experimentally induced anxiety levels reached by subjects listening to no music fn = 30), subjects listening to music selected by the experimenter from the subject's preferred genre or artist listed as relaxing (h = 30), and subjects listening to a specific song they listed as relaxing (n = 30). Subjects consisted of 90 individuals, male and female, randomly assigned to one of the three groups mentioned above. Subjects in either music group filled out a questionnaire prior to participating in the study indicating their preference of music used for relaxation purposes. Subjects in Experimental Group 1 marked their preferred genres and/or artists, and Experimental Group 2 marked specific songs used for relaxation purposes. While the experimenter hypothesized subjects in Experimental Group 2 would show less anxiety than both the control group and Experimental Group 1, there were no significant differences found between the 2 music groups in anxiety levels reached. However, there was a statistically significant difference between the no music control group and both music groups in the anxiety level reached by subjects. Subjects listening to music, both songs chosen by the experimenter and subject selected songs, showed significantly less anxiety than subjects not listening to music

.

For the purposes of this paper, preference will be defined as "the act of choosing, esteeming, or giving advantage to one thing over another through a verbal statement, rating scale response, or choice made from two or more alternatives" (Kuhn, 1980). Music has many styles and forms and elicits equally as many responses from listeners. Attempts to discover the many facets of influence music has on the human body has resulted in measurements of physiological, affective, behavioral, and psychological changes occurring as a result of music interaction (Rawlings, Hodge, Sherr, & Dempsey, 1995; Vanderark & Ely, 1993; Zalanowski, 1990). Many studies have focused on the differences in music perception and preference between people with and without formal music training (Brittin, 1991; Brittin & Sheldon, 1995; Gregory, 1994; Hargreaves & Colman, 1981; Hargreaves, Messerschrriidt, & Rubert, 1980; Morrison & Yeh, 1999). Several of these studies found that musically trained subjects gave significantly higher responses for quality and liking of music examples than did musically untrained subjects. Studies have also found cultural environment to be a primary deciding factor in music preference (Morrison, 1998; Morrison & Yeh, 1999; Siebenaler, 1999). These findings support the importance cultural cues have in forming a music preference decision. While culture had an impact on preference, Siebenaler (1999) found a consistent relationship between the familiarity of the music examples and preference ratings. The numerous factors relating to the formation of music preference by an individual have yielded mixed results for the association between increased familiarity and increased preference (Geringer & Madsen, 1987; Hargreaves, 1984; Hargreaves et al., 1980; Peretz, Gaudreau, & Bonnel, 1998). LeBlanc (1982) presents a model for an interactive theory of music preference with several different levels of variations that account for the formation of different music preferences. Because music preference can change across time, the cycle of preference formation can involve different variables each time a preference decision is made. At the base of the model, cultural environment variables are separated from music input variables by the media variable that is said to draw from both sets to present the listener with a stimulus. This input information must then go through physiological enabling conditions, basic attention, and the listener's current affective state before reaching the personal characteristics of the listener. Once the personal characteristics are added to the input information, the stimulus is processed by the listener's brain. LeBlanc theorizes that the listener can then choose to make an acceptance or rejection preference decision, or choose to explore the stimulus or environment more which would lead to repeated sampling and heightened attention of the stimulus and/or environment. If the listener does choose to accept the stimulus in his/her preference decision, repetition of the stimulus and heightened attention is likely to occur (LeBlanc, 1982). As demonstrated through this model, making a music preference decision is a complex and involved process that varies through time for each individual.

      When selecting music for patients or clients, the preference of the patient/client needs to be considered. Many times music used to reduce anxiety and increase relaxation is slow in tempo and instrumental (no lyrics). However, using LeBlanc's theory of formation of music preference would lead to the assumption that if a person has a history of negative association with this type of slow music, then as results from one study indicate, "that person would not be able to relax with this music, and might even become more tense than he or she would have without music" (Stratton & Zalanowski, 1984, p. 190). The most important factor relating to a client being able to relax is how much the client likes the music (Stratton & Zalanowski, 1984; Thaut & Davis, 1993). Therefore, the personal association each person has with the music played in a therapeutic setting should be taken into consideration. Is the preferred genre a close enough personal association or would specific songs a person has a previous association with offer the best possible therapeutic outcome? This study hypothesizes that there will be no significant difference between any groups in the induced anxiety level differences from pre to post tests. Second, this study hypothesizes that there will be no difference between the two music groups in the induced anxiety level differences between pre and posttests.

     The amount of pain a patient is experiencing can be directly influenced by the anxiety level of the patient. Research has found the more anxious a patient is, the higher perception of pain the patient will experience, causing the perception of pain to continually rise due to increasing anxiety levels about the pain (Andreasen, Noyes, Hartford, Brodland, & Proctor, 1972; Brown, 1990; Flaherty & Fitzpatrick, 1978; Langer, Janis, & Wolfer, 1975; Weisenberg, Wolf, & Raphaeli, 1984). Anxiety levels can be increased or decreased by each individual's stress response. Stress responses are formulated from perceived stressors and influenced by individual personality traits, coping styles and strategies, perceived self-efficacy, and past experiences and expectations (Skodol, 1998; Steelman, 1990). Therefore, an environmental stressor might evoke a high level of anxiety in one patient but a low level in another.

When a stressor of any kind exceeds the individual tolerance level, the generalized stress response is activated. The stress system and response is meant to be activated for a short period of time as a survival mechanism for adjusting the body to homeostasis. Constant or chronic stress however can lead to a stress system dysregulation and ultimately several different disease states. The theory of the body's natural reaction to stress to return to a balanced state can be dated back to Hippocrates who described a healthy body to be in a harmonious balance of the elements and a body affected by disease as one experiencing a systematic disharmony of the elements (Chrousos & Gold, 1992). Individuals in a continually stressful environment are at risk of a stress system dysregulation that could inhibit recovery and healing.

     Several studies have shown music to reduce anxiety, decrease patient's recovery period, decrease length of hospitalization, lower the amount of medication required during recovery, and help confusion to be eliminated (Bolwerk, 1990; Caine, 1991; Chetta, 1980; Hammer, 1996; Kaempf & Amodei, 1989; Liebman & MacLaren, 1991; Pickrell, Metzger, Wilde, Broadbent, & Edwards, 1950; Rider, Floyd, & Kirkpatrick, 1985; Robb, Nichols, Rutan, Bishop, & Parker, 1995; Sanderson, 1986; White, 1992). However, Elliot (1994) found no significant differences in anxiety reduction for patients in a coronary care unit when subjects did not choose the music and listened to light instrumental classical music. While all of the studies mentioned above use "relaxing music", each study varies in the actual artist or genre with selections chosen by the experimenter.

Mixed results have been found in studies allowing patients to choose from an offered selection of music. Anxiety was decreased significantly for subjects in several studies when given a choice of general genres but not specific artists or songs (Finlaw, 1997; Steelman, 1990; Winter, Paskin, & Baker, 1994). A more detailed preference choice was given in a study conducted by Palakanis, DeNobile, Sweeney, and Blankenship (1994) giving subjects undergoing flexible sigmoidpscopy a choice of 20 cassettes representing five different genres. Results from this study showed significantly decreased mean arterial blood pressure (p< .001), anxiety scores on the STAI

(p< .002), and heart rate change (p< .003) in experimental subjects (Palakanis et al., 1994).

Strauser (1997) researched anxiety reduction of patients receiving chiropractic intervention and found no significant difference between groups. Patients in one group chose music from eightgenre tapes, but the experimenter chose the songs on each genre tape (Strauser, 1997). Another study allowing subjects to choose from one of five soothing music tapes also yielded no significant

differences in anxiety reduction between subjects recovering from coronary artery bypass surgery (Barnason, Zimmerman, & Nieveen, 1995). In a study by Winter et al. (1994) several patients refused to participate because they did not like any of the genre choices available. White (1992) also found a subject who expressed that he thought the music was depressing and showed very negative

emotions concerning the music used for the study.

Leite, Seabra, Sartori, and Andreatini (1999) found a way to successfully induce experimental anxiety. Administering a modified Stroop color word test with limited time and all spoken errors signaled by a buzz while subjects see a live projection of themselves on a television screen was tested and proven to induce experimental anxiety. In this modified Stroop color word test, the subjects must say the ink color of each of 100 words while the written word is a contrasting color name. The contrasting messages sent to the brain cause some of the induced anxiety. This model was reproduced for this study to induce experimental anxiety in all subjects.

The present study sought to determine if there is a difference in induced anxiety levels reached by subjects listening to no music, subjects listening to music selected by the experimenter from the subject's preferred genre or artist listed as relaxing, and subjects listening to a specific song they listed as relaxing. Data were collected at three intervals: pretest, during intervention, and posttest for two experimental groups and one control group.

Subjects were randomly assigned to one of the three groups mentioned above prior to any involvement in the study. Dependent variables were the Visual Analog Scale (VAS) where subjects

marked on a 100 mm line to indicate their current level of anxiety with one end of the line being zero anxiety and the opposite end maximal anxiety (Kindler, Harms, Amsler, Ihde-Scholl, & Scheidegger, 2000) and State Trait Anxiety Inventory (STAI) where subjects answered 20 questions concerning state anxiety and 20 questions about trait anxiety (Spielberger, 1983). The trait portion was administered only at the beginning of the experiment immediately prior to the experimentally induced anxiety task due to the nature of trait anxiety being static across time while the state portion was administered at each of three measuring points; pretest, during intervention, and posttest due to state anxiety changing within different situations. Subjects consisted of 90 male and female nonmusic majors either currently in college or having a college degree. Anxiety was induced via the timed Stroop color word test with errors signaled by a buzz with subjects seeing a live projection of him/her on a television screen.

Procedure

Subjects in either experimental music group filled out a questionnaire described below prior to the experimentally induced anxiety portion of the study, indicating their preference of music used for relaxation and motivation purposes. Subjects were asked to list motivation preferences so as not to know the intent of the study beforehand. Subjects in Experimental Group 1 (n = 30) marked their preferred genres and/or artists for relaxation and motivational purposes on a questionnaire listing 21 different genres and 182 artists within the appropriate genre, and a category of "other" for subjects to freely write any genre or artist not on the list. Experimental Group 2 (n = 30) marked specific songs preferred for relaxation and motivation purposes on a questionnaire with the same genres and artists as the genre group but with the addition of 1,799 specific songs listed under the appropriate artist, and a category of "other" for subjects to freely write any specific song not on the list. After the questionnaires were completed, the experimenter chose a song to play from the relaxation selections made by the subjects. Subjects were able to check off as many preferred selections as desired in each category and were given as much time as necessary to complete the questionnaire. Questionnaires were developed from a previous study conducted by the experimenter where subjects listed preferred artists within popular genres (Waiworth, 2000). Subjects listing a specific song were assumed to be making a verbal preference and therefore having had previous enjoyment from that specific stimulus (Price, 1986). Subjects in the control group (n = 30) did not fill out a music questionnaire and received no music during the experiment.

ss         df              MS          F                  P

Between groups      462.200       2             231.100      2.666.        075

Within groups      7,542.200       87             86.692

Total               125,510.000       90

On the day of the experiment, subjects were given a pretest to assess their current level of anxiety with the Spielberger STAI (Spielberger, 1983) and a Visual Analog Scale (VAS). Subjects in either music group were then given a pair of headphones to listen to a song with the volume set at 52 decibels, a level determined by the experimenter as audible but not distracting from the task. After 30 seconds of listening a television and camera were turned on so subjects could see themselves, with music continuing to play in headphones. Subjects were then handed a modified Stroop color-word test and instructed to say aloud the ink color for each printed word, while the actual word was a contrasting color name. Subjects were informed they would have 2 minutes to complete the test with every error signaled by a buzz (Phaf, Christoffels, Waldorp, & den- Dulk, 1998; Williams, Matthews, & MacLeod, 1996). The modified Stroop color word test timed with each error signaled has been proven to be an effective experimentally anxiety producing stimulus (Leite et al., 1999). Anxiety was measured with the state portion of the STAI and a VAS halfway through the Stroop color-word test (after 50 words) and at its completion (100 words). Control subjects received the same procedure without any music.

Results

Two measurements were used to collect data for subjects. The VAS was used at three different points in the study, pre, during, and post. The state portion of the State Trait Anxiety Inventory was

Table of Means

Groups                                  N                    M

1—Control                                30                   33.60

2—Music—selected genre         30                   35.70

3—Music—selected song           30                   39.10

p=.063.

TABLE 3

ANOVA Results for VAS Scores

ss           df           MS        F           P

Between groups         3,753.954         8        469.244      3.253       .003

Within groups          12,548.793        87       144.239

Total                       16,302.747        95

used at the same three points and the trait portion was administered only at pretest. An analysis of variance of the trait portion of the STAI (p = .05) given only at pretest showed no significant difference in trait scores of subjects (see Tables 1 & 2). Randomization of subjects produced groups with no significant difference in anxiety traits. A two-way Analysis of Variance of VAS scores (p = .05) by groups and point in study showed a significant difference between groups (see Table 3). A Duncan LSD Pairwise Comparison post-hoc analysis (p= .05) showed significance between the control group and the two music groups in the post VAS anxiety levels with the music groups being significantly lower (see Table 4) but no significant difference between the two music groups.

A significant difference was found between pretest scores of the state portion of the STAI between the control group and the two music groups. An analysis of variance with pretest as a co-variate was then conducted (p = .05) and showed no significant difference between post-test scores for all three groups (see Table 5). Groups were significantly different in state anxiety prior to the study and the anxiety induction procedure. The various experimental procedures had no significant effect on subjects' subsequent state as measured by the STAI.

Discussion

The State Trait Anxiety Inventory showed no significant differences between groups for this study. Multiple studies were done by Knowles, Coker, Scott, Cook, and Neville (1996) and resulted

TABLE 4

Duncan LSD Post-hoc Means Analysis of Mean VAS Scores

Group                       Pre       After 50 words               After 100 words

Control                     29.50           39.73                      33.77

Preferred genre         37.50           36.50                      29.27

Preferred song          36.87            34.27                      29.17

Total                         34.62            36.80                      30.73

TABLE 5

ANOVA Co-Variate Results of State Scores

---

Between groups            11.835        1       11.835       .537         .466

Within groups                12.944        1       12.944       .587          .466

Total                             1,940.634   88      22.053

in mean shifts in answers on test and retest of anxiety measurement tests. Retest improvement was found multiple times when reporting anxiety levels on tests including the STAI (Elliot, 1994; Knowles et al., 1996). These findings could explain why a decrease in anxiety was not found when measured by the STAI but was found when measured by the VAS. As mentioned earlier, the VAS has been proven to be a valid measurement tool for measuring anxiety (Kindler et al., 2000). A possible research use of this instrument is in the medical setting where anxiety is high for patients, staff, and visitors and is discussed in more detail in the following section. Kindler et al. (2000) tested the validity of using the VAS measuring preoperative anxiety in a study of 685 patients. Results showed the VAS to be comparable to the state portion of the STAI and a valid measurement tool for assessing anxiety (Kindler et al., 2000). Therefore, results of this study seem to indicate that anxiety is decreased when listening to preferred music, whether the preferred genre/artist or a specific preferred song as measured by a visual analog scale (VAS). There was not a significant difference found between music groups indicating that playing a patient or client preferred genre or artist will reduce anxiety as much as playing a specific preferred song with which a patient or client has a previous association developed.

Generalizing the results of this study to the hospital setting would indicate that using patients' preferred music is an effective way to aid in reducing patient anxiety. Results also indicate that using the patients' preferred genre or artist is as effective as using a patientspecified song. Several different areas in the hospital setting that could benefit from this theory and research include pediatric needle sticks and preoperation, adult preoperation, dialysis, chemotherapy treatment, physical therapy exercises, debridement, and labor and delivery. In all of these areas, starting the music before the anxiety producing stimulus occurs has been found to be the most effective in preventing anxiety levels from rising (Standley, 1986) and this study is consistent with that research. Some medical sites are currently utilizing music therapy in the aforementioned areas.

Hospital anxiety for patients has been researched and documented extensively to determine effects of the environmentally induced stress on the human body, and recovery period. Researchers

have found that patients with higher levels of anxiety have extended stays in the hospital, have higher postoperative distress, and can have negative physiological manifestations including a higher

mortality rate (Kulik, Moore, & Mahler, 1993; Steelman, 1990; Vetter, Cay, Phillip, & Strange, 1977). The source of anxiety has also been researched to try to pinpoint what can be changed to alleviate anxiety for the patients. Concern over the illness, the treatment required, uncertainty of the outcome of the illness, the waiting period before surgery, being paired with a preoperative roommate before surgery, and being administered a general anesthetic have all been identified in research studies as anxiety producing for the patients (Brown, 1990; Kaempf & Amodei, 1989; Kindler et al., 2000; Kulik et al., 1993; Shuldham, Cunningham, Hiscock, & Luscombe,

1995; Williams, 1993; Wilson-Barnett, 1979).

The possibility of expanding music therapy services within the hospital to the emergency room seems warranted since that is the first anxiety-producing situation for many patients entering the hospital for treatment. If music therapy was administered during the first stages of patient hospitalization and patient anxiety was initially lowered, results from this study could implicate that the patient's anxiety would not escalate as high over time. Lowered patient anxiety could result in many positive outcomes for the patient and hospital staff including decreased length of stay, decreased medications such as sedatives and pain relievers, and an overall increased

perception of the hospitalization experience.

References

Andreasen, N.J. C., Noyes, R.,Jr., Hartford, C. E., Brodland, G., & Proctor, S. (1972). Management of emotional reactions in seriously burned adults. New England Journal of Medicine, 286, 65-69.

Barnason, S., Zimmerman, L., & Nieveen.J. (1995). The effects of music interventions on anxiety in the patient after coronary artery bypass grafting. Heart & Lung, 24(2), 124-132.

Bolwerk, C. A. L. (1990). Effects of relaxing music on state anxiety in myocardial infarction

patients. Critical Care Nursing Quarterly, 13(2), 63-72.

Brittin, R. V. (1991). The effect of overtly categorizing music on preference for popular music styles. Journal of Research in Music Education, 39, 143-151.

Brittin, R. V., & Sheldon, D. A. (1995). Comparing continuous versus static measurements

in music listeners' preferences. Journal of Research in Music Education, 43, 36-46.

Brown, S. M. (1990). Quantitative measurement of anxiety in patients undergoing surgery for renal calculus disease. Journal of Advanced Nursing, 15, 962-970.

Caine, J. (1991). The effects of music on the selected stress behaviors, weight, caloric intake, and length of hospital stay of premature and low birth weight neonates in a newborn intensive care unit. Journal of Music Therapy, 28, 180-192.

Chetta, H. D. (1980). The effect of music therapy in reducing fear and anxiety in preoperative pediatric patients. Unpublished master's thesis, The Florida State University, Tallahassee, FL.

Chrousos, G. P., & Gold, P. W. (1992). The concepts of stress and stress system disorders. Journal of the American Medical Association, 267(9), 1244-1252.

Elliot, D. (1994). The effects of music and muscle relaxation on patient anxiety in a coronary care unit. Heart & Lung, 23(1), 27-35.

Finlaw, K. (1997). The effects of music on vital signs, peripheral temperature and selected observable stress behaviors of patients hospitalized in a medical-surgical intensive care unit. Unpublished master's thesis, The Florida State University, Tallahassee, FL.

Flaherty, G., & FitzpatrickJ. (1978). Relaxation technique to increase comfort level of postoperative patients. A preliminary study. Nursing Research, 27, 352-355.

Geringer,J. M., & Madsen, C. K. (1987). Pitch and tempo preferences in recorded opular music. In C. K. Madsen & C. A. Prickett (Eds.), Applications of research in music behavior (pp. 204-212). Tuscaloosa: University of Alabama.

Gregory, D. (1994). Analysis of listening preferences of high school and college musicians. Journal of Research in Music Education, 42, 331-342.

Hammer, S. E. (1996). The effects of guided imagery through music on state and trait anxiety. Journal of Music Therapy, 33, 47-70.

Hargreaves, D.J. (1984). The effects of repetition on liking for music. Journal of Research in Music Education 32, 35-47.

Hargreaves, D. J., & Colman, A. M. (1981). The dimensions of aesthetic reactions to music. Psychology of Music, 9, 15—20.

Hargreaves, D. J., Messerschmidt, P., & Rubert, C. (1980). Musical preference and evaluation. Psychology of Music, 8, 13-18.

Kaempf, G., & Amodei, M. E. (1989). The effect of music on anxiety. AORN Journal 50(1), 112-118.

Kindler, C. H., Harms, C., Amsler, F., Ihde-Scholl, T, & Scheidegger, D. (2000). The visual analog scale allows effective measurement of preoperative anxiety and detection of patients' anesthetic concerns. Anesthesia and Analgesia, 90(3), 706-712.

Knowles, E. S., Coker, M. C., Scott, R. A., Cook, D. A., & Neville, J. W. (1996). Measurement-

induced improvement in anxiety: Mean shifts with repeated assessment. Journal of Personality and Social Psychology, 71(2), 352-363.

Kuhn, T. L. (1980). Instrumentation for the measurement of music attitudes. Contributions to Music Education, 8, 2-38.

Kulik, J. A., Moore, P. J., & Mahler, H. I. M. (1993). Stress and affiliation: Hospital roommate effects on preoperative anxiety and social interaction. Health Psychology, 12(20), 118-124.

Langer, E., Janis, I., & Wolfer.J. (1975). Reduction of psychological stress in surgical patients. Journal of Experimental and Social Psychology, 11, 155-165.

LeBlanc, A. (1982). An interactive theory of music preference. Journal of Music Therapy, 19, 28-45.

Leite, J. R., Seabra, M. L., Sartori, V. A., & Andreatini, R. (1999). The videorecorded stroop color-word test as a new model of experimentally-induced anxiety. Progressive Neuro-Psychopharmacology and Biological Psychiatry, 23, 809-822.

Liebman, S. S., & MacLaren, A. (1991). The effects of music and relaxation on third trimester anxiety in adolescent pregnancy. Journal of Music Therapy, 28, . 89-100.

Morrison, S. J. (1998). A comparison of preference responses of white and African-American students to musical/visual stimuli. Journal of Research in Music Education, 46, 208-222.

Morrison, S. J., & Yeh, C. S. (1999). Preference responses and use of written descriptors among music and nonmusic majors in the United States, Hong Kong, and the People's Republic of China. Journal of Research in Music Education, 4 7, 5-17.

Palakanis, K. C., DeNobile.J. W., Sweeney, W. B., & Blankenship, C. L. (1994). Effect of music therapy on state anxiety in patients undergoing flexible sigmoidoscopy. Diseases of the Colon and Rectum, 37(5), 478-481.

Peretz, I., Gaudreau, D., & Bonnel, A. (1998). Exposure effects on music preference and recognition. Memory & Cognition, 26, 884-902.

Phaf, R. H., Christoffels, I. K., Waldorp, L. J., & denDulk, P. (1998). Connectionist investigations of individual differences in stroop performance. Perceptual and Motor Skills, 57,899-914.

Pickrell, K. L., Metzger.J. T, WildeJ. N., Broadbent, R. R., & Edwards, B. F. (1950). The use and therapeutic value of music in the hospital and operating room. Plastic and Reconstructive Surgery, 6, 142-152.

Price, H. E. (1986). A proposed glossary for use in affective response literature in music. Journal of Research in Music Education, 34, 151-159.

Rawlings, D., Hodge, M., Sherr, D., & Dempsey, A. (1995). Toughmindedness and preference for musical excerpts, categories and triads. Psychology of Music, 23, 63-80.

Rider, M. S., Floyd,J. W., & Kirkpatrick.J. (1985). The effect of music, imagery, and relaxation on adrenal corticosteroids and the re-entrainment of circadian rhythms. Journal of Music Therapy, 22, 46-58.

Robb, S. L., Nichols, R.J., Rutan, R. L., Bishop, B. L., & Parker,J. C. (1995). The effects of music assisted relaxation on preoperative anxiety. Journal of Music Therapy, 32, 2-21.

Sanderson, S. K. (1986). The effect of music on reducing preoperative anxiety and postoperative anxiety and pain in the recovery room. Unpublished master's thesis, The Florida State University, Tallahassee, FL.

Shuldham, C. M., Cunningham, G., Hiscock, M., & Luscombe, P. (1995). Assessment of anxiety in hospital patients. Journal of Advanced Nursing, 22, 87-93.

Siebenaler, D. J. (1999). Student song preference in the elementary music class. journal of Research in Music Education, 47, 213—223.