Abstract
This study examines the psychological and emotional effects of music listening interventions on stress and anxiety-related symptoms and emotional responses in adolescents 14-18 years of age. The study was done through a thorough preceding examination of similar studies and research papers over music medicine and therapeutic interventions. In addition, a survey that simulates music listening interventions found in music medicine was distributed on a virtual platform. The creation and structure of the survey is based on information according to the methodology, assessments, and findings of other research papers. The survey includes music samples and self-reported questionnaires that observe stress and anxiety (State Trait Anxiety Inventory, Perceived Stress Scale, and Healthy Unhealthy Music Scale). Results from the study revealed that there is a significant change in state anxiety level for participants with a jazz genre survey version (p = 0.038, α 0.05). There were also observed correlations between stress level v HUMS-Healthy scores, where a lower stress level results in a higher HUMS-Healthy score, and music genre preference v state anxiety level, where receiving a survey version of the most preferred genre of music results in a generally more positive and greater STAI change. These findings could encourage further studies into the field of music listening interventions as a therapeutic method and as a potential highly-accessible way of coping with stress and anxiety on a daily basis.
Introduction
Understanding the impacts of music on the mental health of adolescents could bring benefits to the community. With the onset of the COVID-19 pandemic, many aspects of healthcare, including therapy involving music interventions, have moved to a virtual setting. Virtual therapy is also becoming more widely used with the increasing accessibility to technology and the internet. Using music as a way to achieve self-relaxation and other benefits is common and popular. At the same time, mental health—especially for adolescents—has called more attention to the development of interventions because of its significant impact on social functioning (Porter et al., 2016). The purpose of this study is to conduct a survey involving music listening interventions and observing the emotional responses of adolescents in order to potentially establish a relationship—possibly beneficial—between the music listening and resulting mental health.
Up to 20% of adolescents in the U.S. will experience a depressive episode by the time they are 18 years old. Adolescent depression and anxiety commonly co-occur depending on the physiological, psychological, and social changes they experience (Porter et al., 2016). Many of these changes are a result of similar environments that most adolescents share, such as a school setting or within a friend group. An individual’s experience of emotions is linked to their social environment, and many adolescents live in similar environments, thus resulting in similar impacts on emotional response. With the aspect of music, any significant effect of music on the emotional response of an individual requires consideration and observation of the person’s social surroundings (Sloboda et al., 2001). Music has also been shown to reduce negative emotions and feelings, including state anxiety, and increase positive emotions and feelings (de Witte et al., 2020).
The concept of music therapy overshadows the more specific music interventions and music medicine concepts. Music therapy involves the use of specific aspects of music with a music therapist. The defined relationship between the patient and therapist differentiates music therapy from other music interventions. Active music interventions involve the individual actively engaging with the music through means such as improvisation, music composition, or vocalizing and singing (de Witte et al., 2020). In receptive interventions, the patient responds to the music rather than actively making music; both active and receptive interventions consider specific musical components of the music, including tempo, rhythm, dynamics, and melody (de Witte et al., 2020). Most people who use music for purposes involving self-benefit or pleasure are not engaging in music therapy, but rather music medicine, and more likely receptive music interventions.
This present research studies the emotional effects of music listening interventions on adolescents using receptive music interventions. It contributes to the smaller number of studies that focus on adolescents by targeting individuals 14 to 18 years of age.
Hypotheses
H0A: A participant who completes an experimental survey version(music stimulus) will not have a significant change in state anxiety level compared to a participant who receives a control survey version(non music stimulus).
H1A: A participant who completes an experimental survey version(music stimulus) will have a significant change in state anxiety level compared to a participant who receives a control survey version(non music stimulus).
H0B: Among the participants who completed experimental versions of the survey, there will not be a significant change in state anxiety levels between the categories of key, genre, and tempo.
H1B: Among the participants who completed experimental versions of the survey, there will be a significant change in state anxiety levels between the categories of key, genre, and tempo.
Materials & Methods
Participants
The purpose of the study was to mainly focus on observing the emotional responses in adolescents, so the age range was between 14 to 18 years of age in order to obtain results from individuals who are exposed to similar social changes and communities. The population of interest thus is mainly high school students. Most participants resided in the Dallas Fortworth area Participants were not compensated for their completion of the survey. Participants were told that their answers from their survey would be used for a research study, and that they would remain anonymous and confidential throughout the entire study.
The average age of the 40 participants was 15.8 years. 12.5% of participants were freshmen, 5% were sophomores, 65% were juniors, and 17.5% were seniors. 70% of participants were female, 27.5% were male, and 2.5% marked “other” for gender.
Materials
The survey was created and taken on the voice and video feedback platform Phonic (https://www.phonic.ai/), which allowed for the successful inclusion of audio files into questions. The online platform required participants to utilize an electronic device, most frequently a desktop or laptop, and the audio portion of the survey required a sound output device, most commonly speakers and headphones. Based on the research done prior to this study, the three following assessments were selected to measure the levels of stress and anxiety: State Anxiety version of the State Trait Anxiety Inventory (STAI), Perceived Stress Scale (PSS), and Healthy Unhealthy Music Scale (HUMS).
Study Design
With restrictions to online distribution of the survey and limited access to advanced technology and research primarily due to the COVID-19 pandemic, the survey was determined to be online. This decision was made in order to achieve the most optimal method of obtaining responses. The survey remained anonymous and confidential for the respondents, and was sent online to each individual participant using participants’ emails provided through a Google form. Participants were instructed in the email to complete the survey within 48 hours and given the survey link.
The purpose of the study was to observe music and non-music related effects on adolescents, and thus the survey was composed of a control and experimental groups: the control is composed of white noise and pink noise, both of which are regularly used as music listening interventions for the purpose of obtaining therapeutic effects or self-relaxation (Lu et al., 2020). The experimental variable consists of music with three specific observable attributes.
In the experimental versions of the survey, three major music characteristics were taken into consideration and included during the process of developing the survey: key, genre, and tempo (measured in beats per minute).
The key of a musical composition denotes a scale that creates the foundation for the creation of the composition, and is present in all three genres that were included in the survey.
Three genres were selected based on popularity and use in past studies of the topic of music interventions (de Witte et al., 2020). Classical music was observed across several of the studies and papers used during the pre-research process. Jazz was included as one of the most-preferred genres rated by the general public. Hip hop was included to extend from a previous study that observed therapeutic effects of hip hop lyrics (Tyson, 2002).
Music at a slower tempo(with steady rhythm) was shown to potentially provide stress reduction (de Witte et al., 2020). Music with an identified tempo range was found to have larger effects on stress-related outcomes (de Witte et al., 2019). Thus, there were three designated categories for tempo: slow(60-76 bpm), moderate (76-120 bpm), and fast (> 120 bpm). The parameters of these three categories were based on the most frequently-used terms in classical music to denote tempo: adagio – slow, moderato – moderate, allegro – fast. Studies concerning music listening interventions also utilized specific tempo ranges to categorize the composition used (de Witte et al., 2019).
A within subjects design was used; 20 versions of the survey were created, each with three music compositions of the same key, tempo range, and genre from the categories described above.
Table 1: Individual survey versions’ music compositions(experimental)
The surveys were identified with letters to represent key, genre, and tempo; key was labeled with either a lowercase ‘m’ or uppercase ‘M’ (minor, major respectively); genre was labeled with a ‘c’, ‘j’, or ‘h’(classical, jazz, hip hop respectively); tempo was labeled with ‘s’, ‘mo’, or ‘f’(slow, moderate, fast respectively).
White noise and pink noise consist of random distributions of frequencies as opposed to intentionally spaced frequencies that compose music. The control variable consisted of white noise and pink noise in order to compare effects of musical and nonmusical acoustic stimulation on emotional response. Sounds that fall under the category of white noise and/or pink noise lack the typical components of music, such as rhythm and melody (Thoma et al., 2013), but are still acoustic stimuluses, and thus were used as the control in this study.
Table 2: Control variable audio sample descriptions
All twenty versions of the survey follow the same structural format. The first questions obtain general information from the respondent: age, gender, grade level, race and ethnicity, and musical experience. The participant was asked to complete STAI, PSS, and HUMS questionnaires respectively, before being asked to listen to the audio clips. Participants were then given instructions regarding the three following audio questions.
Phonic survey – instructions given to participants before providing audio stimuli
The same STAI-S questionnaire was given to the respondent to fill out after the three audio questions. The survey included a question regarding the individual’s preference for each of the three genres included in the study (classical, jazz, hip hop), and a final question about the sound output device used.
Phonic survey – Music Preference questionnaire (Likert scale 1-5)
The State Trait Anxiety Inventory is used to assess state anxiety and trait anxiety (STAI-S; Spielberg, 1989). The state version, STAI-S, of the questionnaire is used in studies measuring stress-related outcomes (de Witte et al., 2019). The 20-question State-Anxiety questionnaires were completed before and after providing the audio samples in the survey in order to measure for possible changes as a result of the acoustic stimulation, as seen used in the study by Thoma et al.. Each of the 20 statements are Likert-scale based with a range of 1 to 4(1 – not at all, 2 – a little, 3 – somewhat, 4 – very much so). Before and after results, measured on a scale of 20(low anxiety level) to 80(high anxiety level), were compared to analyze the impact of the acoustic stimulation on state-anxiety levels.
The self-report questionnaire, Perceived Stress Scale(PSS), was used to measure the level of stress of participants before listening to the audio samples (Cohen et al., 1983). The PSS score can range from 0 to 40; scores ranging from 0-13 would be considered low stress, 14-26 would be considered moderate stress, and 27-40 would be considered high perceived stress. The scale ranged from 0 to 4 on a Likert-scale(0 – never, 1 – rarely, 2 – sometimes, 3 – often, 4 always). The scores were calculated for each participant to compare their levels of stress.
The Healthy-Unhealthy Music Scale(HUMS) was developed to address the engagement of music as an indicator of proneness to adolescent depression (Saarikallio et al., 2015). The scale can be used to observe music-based emotion regulation and healthy and unhealthy music use (Silverman, M. J., 2019). The scale consists of 13 items divided into subscales of healthy and unhealthy on a Likert-scale of 5 items (never, rarely, sometimes, often, always), which were calculated for each participant in this study and compared to each other and the other participants.
Results
Data Collection
The STAI-S questionnaire was provided first out of the three questionnaires(STAI-S, PSS, and HUMS) used in the survey prior to the audio questions. Scores were calculated by adding each number provided by the participant for all 20 questions. The number for questions regarding a positive emotion or description was recorded on a reversed 1-to-4 Likert scale. Scores for the before and after STAI-S questionnaires were compared by subtracting the after-score by the before-score. A positive difference signifies a positive impact of the audio stimulus on state-anxiety.
Figure 1: Average STAI-S score change per category
The Perceived Stress Scale (PSS) questionnaire was provided in the question following the STAI-S questionnaire. The number for questions pertaining to a positive description or idea was recorded on a reversed 0-to-4 Likert scale. Scores for the PSS were then calculated by the number provided by the participant per question. Scores from 0-13 are considered a low stress level, 14-26 are considered moderate stress, and 27-40 are considered high perceived stress. According to results from the PSS questionnaire, 15% of participants were found to have a high level of stress, 65% had a moderate level of stress, and 20% had a high level of stress.
Figure 2: Percent of participants per stress level (low stress: 0-13, moderate stress, 14-26, high stress: 27-40)
The Healthy-Unhealthy Music Scale questionnaire was asked after the PSS. The questions were divided into HUMS Healthy and HUMS Unhealthy based on their positive or negative description of music’s role in the participant’s life. The HUMS Healthy is expected to positively correlate with indicators of good mental health, and HUMS Unhealthy is expected to correlate negatively with indicators of good mental health (Saarikallio et al., 2015). Participants’ answers were scored on a Likert scale from 1 to 5 (1-never, 2-rarely, 3-sometimes, 4-often, 5-always). HUMS Healthy and Unhealthy scores were calculated for each individual. Comparisons between the scores for HUMS Healthy and HUMS Unhealthy showed that 85% of participants had a higher HUMS Healthy score than HUMS Unhealthy score, 7.5% of participants had the same HUMS Healthy and HUMS UNhealthy score, and 7.5% of participants had a lower HUMS Healthy score than HUMS Unhealthy score.
Figure 3: HUMS Healthy and Unhealthy scores for participants with classical genre category survey versions
Figure 4: HUMS Healthy and Unhealthy scores for participants with jazz genre category survey versions
Figure 5: HUMS Healthy and Unhealthy scores for participants with hip hop genre category survey versions
Figure 6: HUMS Healthy and Unhealthy scores for participants with control(white noise, pink noise) survey versions
The z-score was calculated for the experimental versions and control versions comparison, and t-scores were calculated for comparisons between control versions and major, minor, classical, jazz, hip hop, slow tempo, moderate tempo, fast tempo experimental versions. p-values were found for all nine categories.
Table 1: z-score and t-score statistics for comparisons of STAI change between music component category and control versions. Statistically significant p-values: **p ≤ 0.05
The experimental versions had a sample size of 36, and its significance compared to the control versions showed a z-score of -1.439, and p-value of 0.150. Eight other comparisons were made between each music component included in the study (key: major & minor, genre: classical, jazz, hip hop, tempo: slow, moderate, fast). The sample sizes for the experimental versions of each music component were below 30, so t-score was used and p-values were calculated for each one. A confidence level of 95% was used; jazz versions compared to the control was the only comparison out of the eight that had a p-value below 0.05. On the other hand, major, minor, classical, hip hop, slow, moderate, and fast versions did not have a significant impact on the change of STAI score (p = 0.232, p = 0.488, p = 0.313, p = 0.245, p = 0.899, p = 0.302, p = 0.295, respectively). The t-score for hip hop versus control was the only one out of the comparisons that was negative because of having a smaller sample mean than population mean(mean of control STAI-changes).
Discussion
The purpose of this study was to find potential changes in stress and anxiety levels as a result of specific music components. We found a significant difference between jazz genre versions and the control versions (p = 0.038), therefore H0B can be rejected. The t-score for jazz versions versus control versions was the largest out of all the comparisons because the mean of STAI-change of the jazz versions was the greatest compared to the other categories. We did not find significant differences between the other experimental variables and the control. There was no significant STAI change in comparing all the experimental versions with the control, thus we fail to reject H0A.
The results from the Perceived Stress Scale questionnaire show that all of the individuals with a jazz version of the survey displayed either a moderate or low level of stress, while some individuals with classical or hip hop versions displayed a high level of stress. We also observed that individuals with low levels of stress had either a higher HUMS-Healthy score than HUMS-Unhealthy score, or the same score for both. Participants with the greatest positive difference between HUMS-Healthy and HUMS-Unhealthy scores also had low levels of stress (M-c-mo: PSS = low, H-Healthy = 25, H-Unhealthy = 8). This suggests that there is a correlation between lower levels of stress and a higher H-Healthy score than H-Unhealthy. There did not seem to be a correlation between HUMS scores and participants who displayed moderate or high levels of stress.
Out of the twenty participants who received survey versions not of their preferred genre out of the three tested genres, five participants displayed a negative STAI difference (25%). And out of the sixteen participants who received a survey version with their most preferred genre, two of them displayed a negative STAI difference (12.5%). This may signify a correlation between state anxiety levels and the selected genre of music based on an individual’s preference. The majority of participants (87.5%) indicated that they had musical experience with a musical instrument sometime in the past and/or present day. However, whether the participants had musical experience or not, both displayed negative and positive STAI differences, so there was no observable correlation between musical experience and state anxiety levels.
Limitations
The research process has identifiable limitations that need to be considered. The sample size was limited to 40 individuals due to the reduction of face-to-face social interaction because of current restrictions from the COVID-19 pandemic, and individuals were given the survey online to take on their own time. Thus, the environment in which each participant took the survey likely varies and could potentially have affected responses. Because there were three experimental variables tested in this study, the number of control samples was very low, which can inaccurately represent the results of the control. This study ideally should have had thirty-six control responses in order to balance the number of experimental responses. Participants were also located primarily in the Dallas Fortworth area, so the findings from this study cannot be generalized for adolescents in other regions of the U.S. or beyond. The selection of music stimuli was based on previous studies involving the use of music listening interventions and also general lists of music therapy song repertoire (University of Kansas Music Therapy Song Repertoire Resource List), and was done based on the three components of music that were observed. As a result, the music selections are likely limited and are not able to fully encompass each music characteristic. It is suggested to take other music components into consideration in future studies and increase the amount and variation of music stimuli given to participants. This study was also limited to a total of three scales to analyze participants’ responses. Therefore, it is suggested that replication studies include other scales that can contribute to the observation of stress and anxiety levels in order to observe potentially more defined data.
Implications
A significant implication of this study is that the use of music listening interventions can affect changes in stress and anxiety levels in adolescents. The stress that the rigorous environment in which adolescents live in today could be lessened with the knowledge of basic effects of music interventions and potential benefits they could bring. The concept and findings from this study could encourage further research in the scope of music interventions and their possible benefits to the mental health of adolescents. Gaining more research in the field of music medicine and interventions can help make music a more established part of mental health care and increase its accessibility to cope with stress and anxiety levels.
References
Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24, 385-396.
Hense, C., Silverman, M. J., & McFerran, K. S. (2018). Using the Healthy-Unhealthy Uses of Music Scale as a Single-Session Music Therapy Intervention on an Acute Youth Mental Health Inpatient Unit. Music Therapy Perspectives. Published. https://doi.org/10.1093/mtp/miy013
Leubner, D., & Hinterberger, T. (2017). Reviewing the Effectiveness of Music Interventions in Treating Depression. Frontiers in Psychology, 8. https://doi.org/10.3389/fpsyg.2017.01109
de Witte, M., Pinho, A. D. S., Stams, G. J., Moonen, X., Bos, A. E., & van Hooren, S. (2020). Music therapy for stress reduction: a systematic review and meta-analysis. Health Psychology Review, 1–26. https://doi.org/10.1080/17437199.2020.1846580
de Witte, M., Spruit, A., van Hooren, S., Moonen, X., & Stams, G. J. (2019). Effects of music interventions on stress-related outcomes: a systematic review and two meta-analyses. Health Psychology Review, 14(2), 294–324. https://doi.org/10.1080/17437199.2019.1627897
Liu Y, Liu G, Wei D, Li Q, Yuan G, Wu S, Wang G and Zhao X (2018) Effects of Musical Tempo on Musicians’ and Non-musicians’ Emotional Experience When Listening to Music. Front. Psychol. 9:2118. doi: 10.3389/fpsyg.2018.02118
Lu, S. Y., Huang, Y. H., & Lin, K. Y. (2020). Spectral Content (colour) of Noise Exposure Affects Work Efficiency. Noise & Health. Published. https://doi.org/10.4103/nah.NAH_61_18
Nater, U. (2003). Music Preference Questionnaire (MCQ-R). https://www.musicandhealthlab.com/app/download/13808949035/MPQ-R_English+version.pdf?t=1515410609
Porter, S., McConnell, T., McLaughlin, K., Lynn, F., Cardwell, C., Braiden, H. J., Boylan, J., & Holmes, V. (2016). Music therapy for children and adolescents with behavioural and emotional problems: a randomised controlled trial. Journal of Child Psychology and Psychiatry, 58(5), 586–594. https://doi.org/10.1111/jcpp.12656
Saarikallio, S., Gold, C., & McFerran, K. (2015). Development and validation of the Healthy-Unhealthy Music Scale. Child and Adolescent Mental Health, 20(4). https://doi.org/10.1111/camh.12109
Sharma, M. (2015). Expanding Horizons Of Academically Stressed Adolescents Through Music Therapy By Enhancing Perceived Social Support And Academic Resilience. Prestige International Journal of Management & IT – Sanchayan, 04(01), 1–12. https://doi.org/10.37922/pijmit.2015.v04i01.007
Silverman, M. J. (2019). Music-based emotion regulation and healthy and unhealthy music use predict coping strategies in adults with substance use disorder: A cross-sectional study. Psychology of Music, 49(3), 333–350. https://doi.org/10.1177/0305735619854529
Sloboda, J. A., & O’Neill, S. A. (2001). Emotions in everyday listening to music. In P. N. Juslin & J. A. Sloboda (Eds.), Music and emotion: Theory and research (pp. 415–429). Oxford University Press.
Spielberger, C. D. (1989). State-Trait Anxiety Inventory: Bibliography (2nd ed.). Palo Alto, CA: Consulting Psychologists Press.
Summer, L. (1994). Considering Classical Music for Use in Psychiatric Music Therapy. Music Therapy Perspectives, 12(2), 130–133. https://doi.org/10.1093/mtp/12.2.130
Thoma MV, La Marca R, Brönnimann R, Finkel L, Ehlert U, Nater UM (2013) The Effect of Music on the Human Stress Response. PLoS ONE 8(8): e70156. https://doi.org/10.1371/journal.pone.0070156
Tyson, E. H. (2002). Hip Hop Therapy: An Exploratory Study of a Rap Music Intervention with At-Risk and Delinquent Youth. Journal of Poetry Therapy, 15(3).University of Kansas Music Therapy Song Repertoire Resource List [Music therapy song repertoire resource list acts as a reference for music therapy students as they work to develop a diverse repertoire of songs for use in their musical, pre-clinical, and clinical work.]. (2021). University of Kansas.
Youth Medical Journal 