IntroductionBackground

Sound plays a fundamental role in human life, influencing communication, emotional regulation, and environmental awareness. Beyond its basic functions in speech and music, sound can modulate psychological and physiological states [,]. Auditory stimuli, including natural sounds, music, and human voices, have been shown to evoke a range of emotional responses and affect stress levels []. Recent research has explored the therapeutic potential of sound-based interventions, particularly in managing stress and promoting relaxation []. Much evidence documented a sound integrative impact on the psychoemotional and physiological outcomes, making it helpful for treating stress-related conditions such as pain syndromes or anxiety [-].

Music and Music Therapy

Sound therapy techniques have gained prominence over the past decades, with a significant emphasis on music as a primary form of sound stimulation. One of the early definitions of music describes it as “humanly organized sound” [], but there are several alternative perspectives debating this definition []. For example, Thaut [] proposed that music is a complex, time-ordered, and rule-based sensory language. Similar to language, music constitutes a universal part of all human cultures, made up of individual sounds that are structured and layered []. The healing effects of music on the mind and body resonate with ancient cross-cultural beliefs []. Nowadays, music therapy incorporates various elements of music, such as melody, rhythm, tempo, dynamics, and pitch, along with activities such as songwriting, improvisation, and singing, to promote patients’ physical and mental well-being.

Nonmusical Sound Therapies

Despite the widespread popularity of music, evidence indicates that other sound types, including natural noises, chanting, and speech, can also exert therapeutic effects. For example, listening to poetry has been shown to alleviate symptoms of anxiety and stress [-]. Moreover, in various therapeutic frameworks, the quality of communication is often considered a crucial factor. The effectiveness of treatment hinges not only on the specific technique or theoretical knowledge but also on the establishment of a trusting and intimate rapport between the client and the therapist. Ackerman and Hilsenroth [] indicate that human-rated factors such as warmth, interest, and curiosity, contribute positively to the development of a strong therapeutic alliance. Therapists use different vocal styles depending on the session’s phase, aligning their voices to meet the emotional and therapeutic needs of the interaction []. However, distinguishing the impact of content from that of speech acoustics in poetry or therapeutic conversations presents a challenge. A comprehensive understanding of the factors that contribute to the therapeutic influence of sound remains to be elucidated.

Mechanisms of Sound Therapy

The theoretical foundation of how sound interventions can yield therapeutic change and positive outcomes encompasses multiple levels, with the most prominent factor being the learned cognitive response shaped by cultural context [].

Four-Level Model

Clements-Cortes and Bartel [] proposed a 4-level model, suggesting that sound can trigger a response due to a set of learned associations that vary from person to person. A particular tone or sound can activate meaningful memories and evoke emotions associated with recalled events []. This phenomenon, known as music-evoked autobiographical memory recall, activates related brain networks, potentially inducing shifts in mood and emotional responses []. Other responses to sounds are closely linked to our inborn perceptive patterns. Music, characterized by organized vibrations and rhythms, reflects early learned responses. This forms the foundation of our reactions to music—fast music feels exciting, while slow music feels calming. Similarly, louder sounds are associated with strength and boldness, while quieter sounds seem weaker and softer [].

These principles extend to nonmusical sounds: slower tempo-rhythmical structures, such as a gentle voice or soft breeze, can feel calming, while faster or louder sounds, such as a storm and thunder, can create higher arousal, excitement, or urgency [,]. In addition to inborn and early learned responses, situational emotional learning also shapes sound perception. Bliss-Moreau et al [] demonstrated that voices can gain emotional meaning through past experiences, influencing how quickly participants respond to emotionally charged words based on their previous associations.

Levels 3 and 4 of the model by Clements-Cortes and Bartel [] encompass responses to sound that are not the result of cognitive processing or learning but rather occur at a vibrational, rhythmic structure. Level 3 focuses on neural oscillatory coherence, where neurons fire in synchrony in response to rhythmic stimulation of the senses (auditory, visual, and tactile), which can lead to various beneficial effects. Level 4 suggests that music and sound can activate mechanisms at the cellular level, potentially influencing everything from neurons to bone and blood cells []. Thaut [] indicates that the brain responds to sound rhythms through involvement, where the listener’s mood aligns with the emotional context conveyed in the sound. This occurs due to acoustic resonance, where the brain naturally synchronizes with the rhythm stimuli [].

Role of Context and Common Therapeutic Factors

In practical applications, it is crucial to consider collateral factors that influence the perception of sound. The effectiveness of music as an art therapy method also depends on the surrounding environment and common factors applicable across creative therapies. These factors include being present in the moment, experiencing a predictable environment, feeling personally connected, developing social skills, finding meaning, feeling motivated, experiencing emotional release, and being actively engaged. Music therapy enhances therapeutic alliance and group processes through playful interactions, shared experiences, musical attunement, synchronicity, and dialogue. Musical engagement also modulates one’s sense of time and space, fostering a state of flow or providing distraction from stress-inducing thoughts []. Another framework for understanding how any therapy, including music therapy, facilitates positive changes is the mediator and moderator model [,]. This model considers external factors that affect the strength or direction of the relationship between treatment and the outcome. Moderators may include client or therapist characteristics (eg, gender, ethnicity, and experience), the format of the treatment (individual vs group and in-person vs online), or the treatment frequency (once vs twice a week) [,].

Objective and Contribution of This Review

This scoping review aimed to analyze literature on sound interventions targeting stress response and stress-related conditions in adults in laboratory experiments, clinical trials, and randomized controlled trials (RCTs). We also aimed to identify sound interventions that have not been sufficiently studied and require further investigation.

We incorporated investigations focusing on responses of the hypothalamic-pituitary-adrenal axis and the autonomic nervous system as indicators of stress, supplemented by self-reported data and introspective surveys as markers of emotional stress. Our primary outcome of interest is the neural mechanisms underpinning the therapeutic influence of sound. In addition, we are interested in the comparison of delivery methods and sound sample choices for understanding the therapeutic factors of sound for stress reduction.

This study marks a conceptual shift in research of sound therapeutic effects by moving beyond music-focused frameworks to consider nonmusical acoustic interventions using human voices or environmental sounds. This broader perspective opens new possibilities for stress management strategies. By highlighting how diverse sounds can serve for therapeutic changes, this review encourages future studies to explore nonmusical sounds more rigorously and investigate how personal, cultural, or environmental contexts can interact with various sound types.

Moreover, this shift is expanding sound therapeutic potential to services such as classical therapeutic methods that use speech and conversation as primary tools. For instance, it is possible to consider the impact of vocal acoustics, such as tone or rhythm, within psychotherapy sessions to enhance emotional attunement and therapeutic alliance.

In addition, this concept can become a basis for the new technological development. It is possible to bring into the light, track, and analyze vocal elements in real time, allowing therapists to adapt their approach based on the client’s stress markers. Environmental sound tracking also holds potential for improving well-being, as ambient soundscapes could be monitored and adjusted to promote relaxation for everyday environments. This broader perspective encourages innovative approaches that personalize sound to individual needs.

MethodsOverview

This review follows the registered study protocol published in JMIR Research Protocols []. The protocol outlines the objective to comprehensively map empirical research on sound interventions for stress reduction. It adheres to Joanna Briggs Institute guidelines and follows the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) checklist [] (), with minor adjustments, which will be detailed in the Protocol Adjustments subsection.

The study protocol defines the eligibility criteria for study inclusion, focusing on adult participants, various sound interventions (eg, music, speech, and nature sounds), and stress-related outcomes. The PRISMA-ScR flow diagram () illustrates the study selection process. Guided by the protocol, this scoping review assesses the comparative effects of sound interventions, examines research directions, and identifies methodological challenges in existing literature.

‎Figure 1. PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) flow diagram. It illustrates the study selection process, including the number of records identified, screened, excluded, and ultimately included in the final analysis []. Research Questions

This scoping review was guided by the population, concept, context framework, recommended for constructing clear objectives and eligibility criteria in scoping reviews []. The population, concept, context framework in this study focuses on the population of interest (adults experiencing stress), the concept (therapeutic sound interventions), and the context (various settings in which sound interventions are applied for stress reduction). Through careful refinement, we formulated the following research questions to guide the entire review process:

What are the therapeutic factors of sound in the case of reducing stress response and stress-related conditions in human adults? For instance, it might be rhythm, sentiment, environmental context, personal preferences, or background.What are the physiological responses associated with sound interventions in stressful conditions, as measured by biomedical technologies and devices?Search Strategy and Query String

This scoping review used a systematic search strategy across 4 key medical and interdisciplinary databases: PubMed, Web of Science, Scopus, and PsycINFO (or EBSCOhost). As the neurobiology of music and sound therapy emerged as a separate field in the 1990s, the search strategy covers the research from 1990 till the present day. The search aimed to capture studies addressing the effects of sound interventions, such as music therapy and guided relaxation, on stress reduction in adults. The complete query string formulated for this purpose was as follows: “(stress OR anxiety OR relax*) AND (“sound therapy” OR “music therapy” OR “guided relaxation” OR “guided meditation” OR hypno* OR ASMR OR MBSR) AND (prosody OR song OR poetry OR voice OR paralinguistics OR paralanguage) NOT (children OR infants OR animal OR teen).”

Using the operator “AND” we combined three key fields to comprehensively capture relevant literature (1) mental stress or relaxation indicators: terms such as “stress,” “anxiety,” and “relax*” were included to identify studies focused on both stress-related conditions and relaxation as therapeutic outcomes; (2) sound-based interventions: this field targeted interventions where sound is essential, encompassing “sound therapy,” “music therapy,” “guided relaxation,” “guided meditation,” “hypno*,” “ASMR,” (autonomous sensory meridian response) and “MBSR” (mindfulness-based stress reduction); and (3) voice and paralinguistic elements: we included terms, such as “prosody,” “song,” “poetry,” “voice,” “paralinguistics,” and “paralanguage” to capture studies examining the acoustic characteristics of speech and natural language as therapeutic components. The exclusion criteria were applied using the NOT operator to filter out studies on children, infants, animals, and adolescents, ensuring a focus on adult populations.

This structured approach aimed to capture a comprehensive range of studies, not only focusing explicitly on sound-based therapies but also those involving therapeutic aspects of speech and vocal elements. By broadening the scope in this way, we aimed to explore potential blind spots in sound therapy research, particularly in the therapeutic roles of natural speech and paralinguistic features alongside traditional music-based interventions.

Study Selection

The process of how citations and full-text reports were reviewed, included, and excluded is shown in the PRISMA-ScR flow diagram (). The first author, a clinical psychologist, led all aspects of the review, including literature search, extraction, screening, and data analysis. The second author, a psychology researcher, assisted in the literature search and extraction and performed screening independently. Disagreements were resolved through discussion and consensus with other authors of the research consortium.

A 2-step process was followed for study selection. First, we screened citation titles, abstracts, and keywords, classifying each citation as “include,” “exclude,” “unclear,” or “duplicate.” In the second step, full-text reports for citations marked “include” and “unclear” were reviewed to make a final decision on inclusion or exclusion. Reference management and screening were conducted in Rayyan (Rayyan Systems Inc), a web-based app for systematic review management. The inclusion and exclusion criteria are presented in and . The inclusion criteria were used to select studies for the review, specifying population, intervention type, outcome measures, study design, and publication type. The exclusion criteria outline conditions under which studies were excluded, including participant characteristics, health conditions, intervention types, and methodological limitations. These criteria ensure a focused review on adult populations using diverse passive sound interventions for stress relief, with an emphasis on studies providing measurable physiological data.

Table 1. Inclusion criteria used to select studies for the review, specifying population, intervention type, outcome measures, study design, and publication type.Inclusion criteriaGeneral description and examplesPopulationHuman adults exposed to stress-related conditions, including both clinical and healthy populations. Examples include the following:
Anxiety, reactive depression, crises and emergency cases, grief, loss, basic needs deprivation, burnout, occupational hazards
Experimental stress
Intervention typeAll types of passive sound interventions. Examples include the following:
Listening to instrumental music, songs, poetry, human voices, and nature sounds
Sound-based therapies, including music therapy, acoustic stimulation during relaxation, meditation, and induction of hypnosis
Outcome measurementsStudies that assess stress relief effects through physiological measures (eg, heart rate variability, salivary cortisol, electrodermal activity, neuroimaging markers)
Studies that combine both physiological measurements and self-reported stress levels (eg, Perceived Stress Scale, State Anxiety Inventory, Beck Anxiety Inventory)
Study designRandomized controlled trials
Clinical trials
Laboratory experiments
Publication typePeer-reviewed original research articles published in English
Table 2. Exclusion criteria, including participant characteristics, health conditions, intervention types, and methodological limitations.Exclusion criteriaGeneral description and examplesPopulationParticipants younger than 18 years
Animals
Individuals with hearing disabilities
Health conditionsConditions for which stress is not the original cause. Examples include the following:
Pain syndromes
Psychotic episodes
Chronic neurological disorders, such as Alzheimer or Parkinson disease
Age-related cognitive changes
Autism spectrum disorders
Tinnitus, hyperacusis
Outcome measurementsStudies relying solely on questionnaires or self-reports without physiological measurement data
Intervention typeActive sound interventions, including improvisation, music composition, drumming, singing, poetry writing, or chanting
Mixed interventions that combine sound with unrelated therapeutic interventions, such as aromatherapy or massage
Mono-interventions: studies in which sound (usually an unclearly defined piece of music) represented the only intervention, and the absence of sound represented the comparator
Data Extraction

After initial screening according to the chosen criteria, 2 reviewers independently evaluated the full-text articles. Key information included the following items: (1) authors and year of publication indicating the first author and publication year; (2) country of origin specifying the country where the study was conducted; (3) concept specifying the main idea and hypothesis of the study; (4) population specifying the sample size and population type (eg, nurses, students, and patients with specific diagnoses); (5) study type, for instance, RCTs, clinical trials, or laboratory experiments; (6) intervention type (groups) indicating the type of sound stimulation as described in the article (eg, music, voice, and nature sounds). In most cases, the separation of participants into groups is based on the intervention. This is where we get information about both the types of sound interventions and the participant grouping; (7) outcomes and measurements specifying the target condition or issue that the sound stimulation aims to impact, and how it was measured, including both physiological measures and self-reports of subjective experience (eg, measurements of stress hormones, heart rate variability (HRV), anxiety scale, and emotional state questionnaires); and (8) key findings related to the scoping review questions—indicating the primary results, such as clinical outcomes, body-mind responses, effectiveness, among others.

Thematic Analysis Method

The thematic analysis for this scoping review followed the reflexive thematic analysis framework by Braun and Clarke [], using an inductive approach to identify key therapeutic factors of sound interventions on stress reduction. The analysis began with a familiarization phase, in which the first 2 authors thoroughly reviewed and familiarized themselves with the data extracted from each included study. This initial stage focused on understanding the study concept and intervention types.

Following familiarization, an open coding process was conducted across the entire dataset, centering on essential elements, such as sound characteristics, and individual outcomes. This coding identified basic elements such as music styles, sentiment, basic physical elements (rhythm and frequency), environmental context, and personal context as core factors within the data. Related codes were then grouped into preliminary categories, organizing the themes based on factors influencing the effectiveness of sound interventions. Three broad categories emerged from this process: (1) sound per se, capturing sound characteristics independent of the listener; (2) personal factors, which focused on listener-dependent characteristics, including musical education, cultural background, personal preferences, and stress history; and (3) environmental factors, covering aspects of sound delivery, such as live versus recorded sound, use of headphones or broadcast sound, and interactions with medical staff or researchers during the session.

Within the category of sound per se, which proved to be the most widely researched, 9 subcategories were developed to reflect specific sound types and qualities. The first 3 of them (1-3) related to music, including different styles, sentiment (such as cheerful vs sad music), and elementary characteristics (such as tempo-rhythm), the next 3 (4-6) outlined nonmusical sounds (eg, white noise, nature sounds, and human voices), and 3 others (7-9) related to the comparison of musical and nonmusical sounds. Initially, the category “environmental context” included subcategories such as “live versus recorded sound” and “presence of supportive personnel,” while “personal preferences” encompassed elements such as “music genre” and “cultural background.” At the same time, due to the modest number of articles assigned to these themes, we decided not to overload the chart and did not include these subcategories in the final edition. In addition, it was important for us to add 3 extra categories for outcomes that were out of the main pattern or for negative outcomes. These were mixed samples (“mix”), relaxing effects of the absence of sound (“0”), and negative, stressful sound effects (“–”).

During the thematic refinement stage, categories and subcategories were iteratively adjusted through regular team discussions and codebook updates, enhancing internal coherence and consistency across the themes. As the thematic map took shape, each category was further refined and defined to ensure clarity and distinctiveness. This thematic framework not only provided a structure for categorizing therapeutic factors and mechanisms of sound interventions but also revealed patterns in the literature and highlighted gaps for future research. As shown in the Results section, some categories have little research, suggesting areas for future studies. We found that in most of the studies reviewed, the hypotheses and groupings were based on the type of sound intervention. This means that the groups typically reflect the type of sound being tested. In some cases, there were >2 types of sounds. If the categorization was unclear or debated, we referred to the study’s concept and research question to identify the primary focus of authors. For instance, if a study had 3 groups, “cheerful music,” “sad music,” and “white noise,” we would place it in the “musical sentiment” subcategory as the study’s focus is on musical emotion rather than comparing music to nonmusical sounds.

Protocol Adjustments

This section outlines the adjustments made during the review process compared to the study protocol []. While the overall methodology remained consistent, some procedural differences emerged during implementation. One notable change was the exclusion of gray literature sources, such as Google Scholar, ClinicalTrials.gov, and nonindexed conference proceedings. The final review focused exclusively on peer-reviewed studies to maintain methodological rigor. Many gray literature sources lacked transparency in their methods, making it difficult to assess their reliability. In addition, most did not report detailed physiological stress markers, such as cortisol levels, HRV, or neuroimaging data, which were essential for inclusion in this review. Instead, they often emphasized subjective well-being and relaxation effects, which, while relevant, did not align with the primary focus of this study. Given these challenges, the review prioritized published studies that had undergone formal peer review, ensuring that all findings were based on reliable findings. Despite these adjustments, the core thematic analysis approach, guided by the framework by Braun and Clarke [], remained unchanged. These refinements ensure that the findings accurately reflect the available evidence while maintaining alignment with the original research objectives.

ResultsOverview

The systematic database search obtained 2027 records. After removing duplicates, 1924 records remained for further review. Screening of titles and abstracts led to the exclusion of records that did not meet the chosen criteria, such as unsuitable languages, outcomes, or interventions. Full-text review of the remaining 41 records revealed that 7 did not meet the inclusion criteria due to various reasons, including wrong intervention (n=1, 2%), wrong study design (n=5, 12%) and lack of data on key variables (n=1, 2%). Finally, 34 studies were considered eligible for inclusion in the analysis. These studies were RCTs (22/34, 65%), experiments (9/34, 26%), and pilot studies (3/34, 9%). The search process is outlined in the PRISMA-ScR flowchart (). A summary of the included studies, detailing their country of origin, research concept, participant population, and study design is outlined in .

Table 3. An overview of the included studies.Study and yearCountryConceptPopulationStudy typeUmemura and Honda [], 1998JapanClassical music may promote relaxation by influencing heart rate variability compared to rock music or noise.6 university students, aged between 21 and 26 yearsExperimental studyChafin et al [], 2004United StatesListening to classical music may improve blood pressure recovery after a stressful task compared to silence or other music styles.75 healthy participantsExperimental studyLabbé et al [], 2007United StatesListening to classical or self-selected relaxing music reduces anxiety, anger, and physiological arousal after stress compared to heavy metal music or silence.56 college studentsRCTaUğraş et al [], 2018TurkeyAll types of music can reduce preoperative anxiety, with classical Turkish music being the most effective.180 patients undergoing surgeryRCTGulnahar and Kupeli [], 2020TurkeyListening to music reduces anxiety during dental implant surgery, with Turkish music and classical music potentially being most effective.80 dental implant surgery patients, aged between 40 and 70 yearsProspective, observational RCTPaszkiel et al [], 2020PolandRelaxing music and autonomous sensory meridian response music may reduce stress levels faster than silence or rap music. Rap music may even worsen stress compared to silence.9 healthy female participants, aged 22yearsPilot studyHirokawa and Ohira [], 2003JapanHigh-uplifting and low-uplifting music may have different effects on immune function, neuroendocrine responses, and emotional states after a stressful task.18 Japanese college studentsExperimental studySokhadze [], 2007United StatesBoth pleasant and sad music may improve physiological recovery after stress (viewing stressful images) compared to white noise. White noise may not enhance recovery.29 healthy participantsExperimental studySuda et al [], 2008JapanMajor mode music reduces stress more than minor mode music.10 graduate studentsExperimental studyWiwatwongwana et al [], 2016ThailandMusic may reduce anxiety in patients undergoing cataract surgery. Binaural beat audio may offer additional benefits over music alone.141 patients undergoing cataract surgery under local anesthesiaProspective, double-blind RCTOpartpunyasarn et al [], 2022ThailandBinaural beat audio may reduce anxiety in patients undergoing fiber-optic bronchoscopy compared to plain music or no music.112 patients undergoing fiber-optic bronchoscopyProspective RCTLee-Harris et al [], 2018United KingdomMeditative binaural music may be effective for relaxation, with effects potentially differing by age.30 (15 and 15) participants, aged between 18and 25 years and 50 and 80 yearsRCTGantt et al [], 2017United StatesBinaural beat technology (theta waves) embedded in music may be more effective than music alone in reducing cardiovascular stress response in military personnel with postdeployment stress.74 military service members with postdeployment stressDouble-blind RCTCalamassi et al [], 2022ItalyListening to music at 432 Hz and 440 Hz may reduce anxiety and stress biomarkers compared to a no-music control group.54 emergency nursesDouble-blind RCTSharma et al [], 2021IndiaIndian classical music with incremental tempo and octave variations promotes better anxiety reduction.21 male undergraduate medical studentsCrossover RCTSingh et al [], 2009IndiaBoth music and progressive muscle relaxation are effective in reducing anxiety and dyspnea in hospitalized patients with COPDb. Music may be more effective than PMRc for reducing anxiety.72 hospitalized patients with COPD with recent exacerbationRCTTang et al [], 2009United StatesAudio relaxation programs may be effective for short-term blood pressure reduction in older adults, potentially more effective than listening to Mozart.41 older adultsRCTLin et al [], 2011TaiwanMusic therapy and verbal relaxation are effective in reducing anxiety induced by chemotherapy.98 patients with cancer under chemotherapyRCTLee et al [], 2012GermanyBoth monochord sounds and progressive muscle relaxation can reduce anxiety and improve relaxation during chemotherapy.40 patients with gynecologic cancer under chemotherapyRCTWarth et al [], 2016GermanyLive music therapy may be more effective than a prerecorded mindfulness exercise in improving cardiovascular health for patients who are terminally ill through its influence on the autonomic nervous system.84 patients in palliative careRCTKoehler et al [], 2022GermanyMusic therapy may be more effective than mindfulness in reducing subjective distress in patients in palliative care. Both interventions may reduce stress biomarkers.104 patients in palliative careRCTRadstaak et al [], 2014NetherlandsListening to self-chosen relaxing or happy music after stress may improve mood but delay systolic blood pressure recovery.123 healthy participantsExperimental studyLeardi et al [], 2007ItalyMusic therapy, especially patient-selected music, may reduce stress response during day surgery compared to no music.60 patients undergoing day surgeryRCTMiller et al [], 2010United StatesMusic may influence endothelial function, potentially improving vascular health.10 healthy participants, average age 35.6 yearsCrossover RCT with counterbalancingImbriglio et al [], 2020CanadaGuided music listening with relaxing music or a participant’s favorite music may decrease muscle activity and bruxism episodes in chronic myalgia, while stressful music may increase them.14 women with chronic myalgia and 15 women who are pain freeExperimental studyGelatti et al [], 2020ItalyLive harp music may be more effective than recorded harp music in reducing preoperative stress, fear, heart rate, and blood pressure.46 patients undergoing day surgeryPilot, quasi-experimental studyBro et al [], 2019DenmarkLive music may reduce anxiety during chemotherapy compared to prerecorded music or standard care.143 newly diagnosed patients with lymphomaMulticenter RCTLee et al [], 2011TaiwanBoth headphones and broadcast music can effectively reduce preoperative anxiety in adult patients undergoing surgery.167 patients undergoing surgery without premedicationRCTKumari et al [], 2023IndiaBroadcast and headphone music playing may vary the anxiety-relieving effect for patients awaiting surgery.150 healthy participantsExperimental studyLai et al [], 2012TaiwanMusic intervention with a nurse present is more effective than recorded music alone for improving psychophysiological health in caregivers for patients with cancer.34 female caregivers for patients with cancerCrossover RCTJanelli et al [], 2004United StatesListening to preferred music may reduce negative behaviors in patients who are physically restrained.30 patients who are physically restrained, aged between 65 and 93 yearsPilot studyKang et al [], 2008South KoreaBlocking noise, but not music, reduces bispectral index scores during propofol sedation in noisy operating rooms.63 patients undergoing total knee replacement surgery, aged between 55 and 75 yearsProspective, single-blind RCTTsivian et al [], 2012United StatesMusic with headphones may reduce pain perception and anxiety during prostate biopsy compared to control or headphones alone.88 men undergoing transrectal ultrasound prostate biopsyProspective RCTGingras et al [], 2014AustriaRepetitive drumming with shamanic instructions may affect subjective experiences and cortisol levels compared to instrumental meditation music.39 participants inexperienced in shamanic journeyingExperimental study

aRCT: randomized controlled trial.

bCOPD: chronic obstructive pulmonary disease.

cPMR: progressive muscle relaxation.

Thematic Analysis Results

Thematic analysis of the included studies revealed several key factors contributing to the effectiveness of sound interventions. These factors can be categorized into 3 main categories: sound per se, personal factors, and environmental factors ().

The majority of the identified articles focused on the characteristics of sound per se (22/34, 65%). For example, studies investigating music interventions often explored different music styles (6/34, 18%), music sentiment (3/34, 9%), and physical aspects such as frequency or tempo (6/34, 18%).

Regarding nonmusical sounds within our predefined inclusion criteria, we found publications comparing musical interventions with a variety of voice-assisted interventions (7/34, 21%). However, no studies were found that focused narrowly on nonmusical sounds for stress relief in adults or compared different sound characteristics beyond the concept of musical intervention.

Within the personal factors category, listener characteristics (3/34, 9%) were identified as influencing intervention outcomes. Environmental factors, such as the way of delivery and context of delivery, were also considered (6/34, 18%).

In addition, the analysis identified a small number of studies (1/34, 3%) that explored mixed samples of sounds and a few that shifted their focus from sound to silence (2/34, 6%) or reported negative effects of sound intervention (3/34, 9%).

Information on key findings for comparing the effectiveness of different sound intervention factors is summarized in .

‎Figure 2. A summary of the thematic analysis, categorizing sound interventions into key therapeutic factors, including sound characteristics, personal factors, and environmental influences on stress reduction. The numbers in white circles represent classification labels assigned to categories based on thematic analysis (see the Thematic Analysis Method under Methods for details). Table 4. The effects of sound interventions.Study and yearInterventions (groups)CategoryOutcomes (measurements)Key findingsUmemura and Honda [], 19981HRVa (MWSAb and RSAc components) and subjective comfort levels (psychological evaluation)Classical music reduced variability in HRd components (MWSA and RSA) compared to rest, suggesting a potential relaxation effect. Rock music and noise increased MWSA and decreased RSA, potentially indicating a stress response. Subjectively, classical music was associated with comfort, while rock music and noise were associated with discomfort. Changes in MWSA correlated with comfort levels, suggesting a link between HRV and perceived comfort.Chafin [], 2004Classical music
Jazz music
Pop music
No sound (control)
1SBPe measured before and after the stressful task.Listening to classical music resulted in significantly lower posttask SBP compared to silence. Other music styles did not show significant benefits for blood pressure recovery.Labbé et al [], 2007Classical music
Self-selected relaxing music
Heavy metal music
No sound (control)
1Emotional state (anxiety and anger) and physiological arousal (before and after intervention)Listening to self-selected or classical music significantly reduced negative emotions and physiological arousal compared to heavy metal music or silence.Uğraş et al [], 2018Turkish classic music
Western music
Natural sounds
No sound (control)
1STAI-Sf) for anxiety, SBP, DBPg, HR, and cortisol levels measured before and after music interventionAll music types significantly reduced anxiety (STAI-S), SBP, and cortisol levels compared to before the intervention. Natural sounds reduced DBP, and classical Turkish music showed the most significant reductions in DBP and HR. All music was effective in reducing preoperative anxiety, with classical Turkish music being the most beneficial.Gulnahar and Kupeli [], 2020Turkish classic music
Classical music (Vivaldi)
Slow rock music
No sound (control)
1Pre- and postintervention anxiety using the Corah Dental Anxiety Survey, blood pressure, HR, and oxygen saturationAll music groups showed reduced anxiety compared to controls. Turkish music and classical music were significantly more effective in anxiety reduction compared to slow rock music.Paszkiel et al [], 2020Rap music
Relaxing music
ASMRh music
No sound (control)
1EEGi–alpha wave activity, blood pressure, HR, and subjective stress perception questionnairesRelaxing music and ASMR reduced stress levels faster than silence. Rap music increased stress levels compared to silence.Hirokawa and Ohira [], 2003High-uplifting music
Low-uplifting music
No sound (control)
2S-IgAj level, active NKk cell level, T lymphocyte subsets (CD4+, CD8+, and CD16+), dopamine, norepinephrine, and epinephrine levels (before and after intervention), and emotional state questionnaires (before and after intervention)Results were inconclusive, but trends suggested (1) low-uplifting music may increase feelings of well-being; (2) high-uplifting music may increase norepinephrine and liveliness and decrease depression; and (3) silence may decrease active NK cells. Music classification is important for understanding music’s influence on these responses.Sokhadze [], 2007Subjectively pleasant music
Sad music
White noise
No sound (control)
2EEG—frontal and temporal activity, skin conductance, HR, HRV, facial capillary blood flow, respiration rate (measured before and after stressful images and during recovery interventions)Both pleasant and sad music improved recovery on most measures compared to white noise (HR, respiration rate, and blood flow). White noise did not enhance recovery. Both music types positively influenced cardiovascular and respiratory activity during recovery. The “undoing hypothesis” (positive emotions aid recovery from negative emotions) was partially supported.Suda et al [], 2008Major music
Minor music
No sound (control)
2Salivary cortisol levels (endocrine stress marker) and optical topography (brain activity)Major mode music resulted in lower salivary cortisol levels (stress marker) compared to minor mode music. This suggests music can induce emotional responses such as happiness, potentially linked to stress reduction.Wiwatwongwana et al [], 2016Binaural beat music
Plain music
No sound (control)
3STAI score (before and after intervention), SBP, and HR (measured at admission, surgery start, and 20 min after surgery).Both music with and without binaural beats reduced anxiety (STAI score) and SBP compared to the control group. Binaural beat music showed an additional decrease in HR compared to plain music or the control group. Binaural beat audio embedded in music may have some advantages over music alone for reducing anxiety during cataract surgery.Opartpunyasarn et al [], 2022Binaural beat music
Plain music
No sound (control)
3STAI score (before and after bronchoscopy), blood pressure, HR, and sedative usePatients in the binaural beat group showed a significantly greater reduction in anxiety (STAI score) compared to the plain music and no music groups. They also experienced a decrease in blood pressure but an increase in HR. Binaural beat audio may be effective for reducing anxiety before bronchoscopy.Lee-Harris et al [], 2018Meditative music with binaural beats
Meditative music without binaural beats
Low-arousal classical music
High-arousal classical music
3Self-reported emotional state (arousal and positivity) and physiological arousal (measured but not specified in the abstract).The effect of listening to MBMl was comparable to listening to calm classical music. Younger adults showed a stronger decrease in self-reported arousal with MBM with binaural beats compared to low-arousal classical music. Older adults showed a preference for low-arousal classical music for feeling comforted, followed by MBM. These results suggest age may influence how music affects relaxation.Gantt et al [], 2017Music with embedded theta binaural beats
Music alone
3HRV was measured before and after the intervention to assess changes in sympathetic and parasympathetic nervous system activity. Daily self-reported stress levels were recorded in diaries.Music with binaural beats led to a decrease in low-frequency HRV (indicating reduced sympathetic response) and an increase in high-frequency HRV (indicating increased parasympathetic response) compared to music alone. The binaural beats group also reported lower daily stress levels compared to the music-only group. Overall, music with theta binaural beats showed promise in reducing physiological and psychological stress markers in military personnel with postdeployment stress.Calamassi et al [], 2022Listen to 440 Hz music during break
Listen to 432 Hz music during break
Usual break activities (control)
3STAI-X1, HR, respiratory rate, SBP and DBP, and pain and productivity (Likert scale)All groups showed a reduction in anxiety after the break. Listening to 432 Hz music showed the greatest reduction in anxiety and additionally reduced respiratory rate and SBP. Music at 440 Hz reduced anxiety but showed no significant physiological benefits in this study.Sharma et al [], 2021VMm with incremental tempo or octave changes
SMn without variations
No sound (control)
3Anxiety (Beck Anxiety Inventory and STAI), EEG for brainwave activity, ECGo for HRVSignificant anxiety reduction only in the VM group. VM showed decreased low-frequency brainwaves and midline power (reduced default mode network activity) compared to silence. VM also showed more balanced brain activity compared to SM. HRV remained stable during music interventions. The authors propose VM induces a “controlled mind wandering” state leading to anxiety reduction.Singh et al [], 2009Self-selected music
Audio instructions (PMp)
7STAI (Spielberger inventories), dyspnea, SBP and DBP, pulse rate, and respiratory rate (before and after intervention)Both music and PMR significantly reduced anxiety, dyspnea, SBP, pulse rate, and respiratory rate. However, the music group showed greater reductions in anxiety compared to the PMR group. Music and PMR are effective for reducing anxiety and physiological measures in hospitalized patients with COPDq, with music potentially being more effective for anxiety reduction.Tang et al [], 200912-minute audio relaxation program
12-minute Mozart andante
7Blood pressure (systolic and diastolic) was measured at each intervention session and at 1-month and 3-month follow-upBoth groups showed significant reductions in blood pressure after the intervention. The audio relaxation group showed a greater reduction in SBP than the Mozart group. Blood pressure reductions were not sustained at 1- and 3-month follow-up. Audio relaxation may be more effective than music for short-term blood pressure reduction in older adults.Lin et al [], 20111 hour session of music therapy
30 minutes of guided verbal relaxation
Usual care
7State anxiety (Spielberger STAI), emotional distress (Emotional Visual Analog Scale), and physiological responses (skin temperature, HR, and consciousness level) before and after chemotherapyMusic therapy was more effective than verbal relaxation or usual care in reducing postchemotherapy anxiety and increasing skin temperature. Patients with high initial anxiety benefited more from music therapy than those with normal anxiety. Both interventions showed some effectiveness within 30 minutes.Lee et al [], 20127Spielberger State Anxiety Inventory for Anxiety, questionnaires on physical and psychological states, EEG data before and after interventionBoth the monochord sounds and PMR groups showed significant improvements in anxiety, physical, and psychological state. EEG data showed increased posterior theta and decreased midfrontal beta-2 activity in both groups, indicating relaxation. The monochord sounds group also showed a decrease in alpha band activity compared to PMR. Both interventions were effective for reducing anxiety and improving relaxation, with potentially different underlying neural mechanisms.Warth et al [], 2016Live music therapy
Prerecorded mindfulness exercise
7Vagally mediated HRV and blood volume pulse amplitude were measured over time to assess autonomic nervous system responseBoth groups showed improvements over time, but music therapy led to significantly greater reductions in vascular sympathetic tone (BVP-Ar). This suggests music therapy may be more effective in managing pain and stress symptoms in palliative care. Baseline pain levels influenced patient response, highlighting its importance in treatment planning. Music therapy’s impact may be related to the therapist-patient relationship in addition to the music itself.Koehler et al [], 20223 sessions of music therapy
3 sessions of mindfulness training
7Subjective distress rating scale (before and after intervention), salivary cortisol and alpha-amylase levels (before and after intervention), and HR and HRVMusic therapy showed a greater reduction in subjective distress compared to mindfulness. Both interventions led to reductions in cortisol and HR, but no significant differences between groups were found for these or other stress biomarkers.Radstaak et al [], 2014Self-chosen relaxing music
Self-chosen happy m