The Healing Power of Music: How It Lifts Your Mood and Eases Stress, Anxiety, and Depression

Play C, E and G together on a piano, then move the E down one key to E♭. Most people raised on Western music hear the second chord as darker, although nothing in the sound itself says "sad". That small change holds the whole subject of this article: music moves us through patterns our brains have learned to predict, and researchers can now measure some of what happens when it does.

This guide covers why music feels good, what composers do to move you, why singing together bonds people, and what music can and cannot do for stress, anxiety, depression and dementia, including the work of board-certified music therapists. Along the way you'll find examples to play, a ten-minute routine for calm and the free worksheets that go with them.

One promise up front: music can lift your mood, and music therapy has helped people with depression in clinical trials, but it complements treatment and never replaces it. If you are in crisis or thinking about suicide, call or text 988 in the US or 9-8-8 in Canada, call Samaritans on 116 123 in the UK and Ireland, or find a free helpline for your country at findahelpline.com. There is more on getting help near the end.

Why music feels good: the reward system

Pleasure from music runs through the same brain circuitry that responds to food and other rewards. In a 2001 brain-imaging study, as the chills listeners got from music they had chosen themselves grew stronger, blood flow changed in regions linked to reward and emotion, among them the ventral striatum, the amygdala and the orbitofrontal cortex.

A full house for the pianist Josef Hofmann at the old Metropolitan Opera House in New York, November 28, 1937, seen from behind the stage. The studies in this section measured listening, not playing. Image: public domain (U.S. National Archives), via Wikimedia Commons.

Ten years later, the same Montreal lab measured dopamine itself. Valorie Salimpoor and colleagues asked eight listeners to bring music that reliably gave them chills, then scanned them with PET using a tracer that competes with dopamine for its receptors. Dopamine release in the striatum was greater during that music than during neutral music. A second scan, with fMRI, added the timing: the caudate was more active during the build-up before a chill, the nucleus accumbens at the chill itself. Anticipation, it seems, is rewarding in its own right.

The two moments of the 2011 study, drawn as a sketch: the caudate was more active in the build-up you can feel coming, the nucleus accumbens at the chill. The curve's shape and the band widths are illustration, not measurements.

Brain scans show correlation, not cause. The causal test came in 2019, when Laura Ferreri and colleagues gave 27 healthy volunteers, on three separate days and double-blind, a drug that boosts dopamine (levodopa), one that blocks it (risperidone) and a placebo, and played them music each time. Compared with placebo, levodopa raised pleasure and the money people were willing to spend on the songs, and risperidone lowered both. Dopamine is part of why music feels good, not a bystander.

Two caveats keep this honest. These were small studies of healthy people listening to music they liked: they show how musical pleasure works, not that music treats anything. And people differ. Only 16 of Ferreri's 27 volunteers reported chills in the placebo session, and some healthy people get little or no reward from music, a trait called specific musical anhedonia.

Four sources of musical pleasure covered in this guide: anticipation, surprise, the peak, and making music together.

Expectation and surprise: what composers do to move you

Much of the build-up comes from prediction. Your brain is always guessing what comes next, and how those guesses turn out shapes how much you enjoy the music. In a 2019 study, a computer model rated how predictable each of 80,000 chords from US Billboard pop songs was. Listeners enjoyed a chord most when they were fairly sure what was coming and got a surprise, or when they were unsure and then heard something expected.

The deceptive cadence is that surprise at its plainest. Play C, F, G7 and C: the last chord is the full stop your ear expects, an authentic (or perfect) cadence. Then play C, F, G7 and A minor: the bass rises one step to A instead of to C, and the phrase lands on "the sad twin", as our Cadences in Every Key sheet calls the vi chord. Cadences explains the other endings, and Ear Training: Cadences & Progressions lets you test your ear on all four.

The same three chords, then two endings. Top: G7 to C, the authentic cadence. Bottom: G7 to A minor, the deceptive cadence. Right hand: the chord voicings of our Cadences in C sheet; left hand: each chord's root.

Composers save such moves for the moments that matter. In a classic 1991 survey, the psychologist John Sloboda asked music lovers which passages gave them shivers, tears and other strong physical reactions. The most common triggers, a 2014 review reports, were melodic appoggiaturas, unexpected harmonies, sequences, and chord progressions falling around the circle of fifths to the home chord; later studies added peaks and sudden jumps in loudness and moments of modulation, where the music changes key. Treat this self-report survey as a map of where to look, not a law.

Two of those devices are easy to feel at the piano. An appoggiatura leaps onto a note that clashes with the chord, on the beat, then steps to a chord tone. A suspension holds a note over from the previous chord so that it clashes with the new one, then lets it fall. Our Non-chord tones sheet sums them up as "leap in, step out" and "prepare, hold, resolve", and Suspended chords shows the same 4–3 fall inside a single chord. Play each example slowly and lean into the clash: the release is where the feeling lives.

From our Non-chord tones sheet, with its labels. Left: a 4–3 suspension, the F prepared in the F chord, held into the C chord, then falling to E. Right: an appoggiatura, a leap onto C over the G chord, resolving down to B.

Loudness and drive build tension on a larger scale. In Chopin's Prelude in E minor, the melody spends long stretches sighing between two neighboring notes; then bar 16 is marked stretto ("pressing on"), and bar 17 brings the only forte in the piece, with the melody's highest note over a deep bass. Inside Chopin's Prelude in E Minor follows the whole arc, and The Dynamic Ladder trains the six levels from pp to ff, with a golden rule that fits here: save ff for the moments that deserve it.

Prelude in E minor, Op. 28 No. 4, bars 16–17: stretto, then the only forte in the piece. Notes and marks from the verified score of our Chopin study. The complete score of Chopin's Prelude in E minor is in our sheet music library.

Major, minor, tempo and loudness: the cues you hear

Some emotional cues sit in a single note. Our Major & minor sheet calls it "one note changes the weather": lower the third of a C major chord from E to E♭ and the chord turns minor. Try it with a tune, as the sheet suggests; Ode to Joy works well: play it in C position, then again with every E turned to E♭. In this phrase nothing else changes, because it uses only C, D, E, F and G. Parallel vs Relative Minor shows the whole C minor scale beside C major.

Ode to Joy, phrase 2, in C major and then C minor: every E becomes E♭ and nothing else changes. Beethoven's tune as on our Phrasing sheet, in C position with the rhythm simplified. Play the full piano arrangement of Ode to Joy from our sheet music library.

To see which cues matter most, a 2013 study by Tuomas Eerola, Anders Friberg and Roberto Bresin varied six of them systematically across 200 short examples: mode, tempo, register, loudness, articulation and timbre. Listeners rated how happy, sad, peaceful and scary each one sounded. On average mode mattered most, then tempo. Happy examples were major, fast, high and detached; sad ones minor, slow, low, smooth and soft; peaceful ones major, slow, high, smooth and soft; scary ones minor, low and loud. Notice what was measured: the emotion people heard in the music, not necessarily the one they felt.

The cue profiles listeners heard as happy, sad, peaceful and scary in Eerola, Friberg and Bresin (2013). "No clear effect" means that cue did not move the ratings for that emotion.

You can hear the profiles in one small motif: sad and calm share slow, soft, smooth playing, and mode and register tell them apart. For what tempo does to your heart rate and sleep, see our companion article The Soundtrack of Wellness.

An original motif set to the three cue profiles. Happy: major, fast, high and detached. Sad: minor, slow, low, smooth and soft. Calm: major, slow, high, smooth and soft.

Two findings keep the major–minor story in proportion. Children learn the cues in stages: in a 2001 study, 5-year-olds judged happy and sad excerpts by tempo alone, while 6- to 8-year-olds, like adults, used mode as well. And culture matters. The Mafa, an African people with no exposure to Western music, recognized happy, sad and scared Western excerpts above chance, but in Papua New Guinea, major cadences sounded happier than minor ones in every community except the one with the least exposure to Western-style music. The link between minor and sad is strongly tied to what you grew up hearing.

Use the cues on purpose. In a 2015 study of 123 people, those who often used music to express negative feelings scored higher on anxiety and neuroticism, and a 2017 survey found people with depression more likely to ruminate with sad music in groups. Both are correlations, not proof that the music did harm, but they suggest a habit: notice whether a playlist lifts you or keeps you circling.

Why singing together works: music and social bonding

Music is also social glue, and here the evidence comes mostly from making music rather than listening. Robin Dunbar's group at Oxford tests this with pain threshold, an indirect sign that the brain has released endorphins, its own opioids. In small 2012 field studies, pain thresholds rose after a church meeting full of singing and clapping, a samba drumming session and dance classes, but not after listening to music, and not reliably after the stop-start rehearsals of a choir, a brass band and a small orchestra. Energetic music-making, not the sound alone, seemed to matter.

A singalong at the Neustädter Kirche in Hannover, Germany, December 2011. Photo: Buddi1947, CC BY-SA 3.0, via Wikimedia Commons (cropped).

Singing together also speeds up friendship. When Eiluned Pearce and colleagues followed new adult-education classes for seven months, the singing classes bonded much faster, while the craft and creative-writing classes caught up only by the end: an "ice-breaker effect". In a community choir study, pain thresholds, positive mood and closeness all rose after 90 minutes of singing, both in small groups of 20 to 80 and in a 232-voice "megachoir" of people who did not know each other, where closeness grew the most. Moving in time seems to be part of it: in a silent-disco experiment, people who danced in synchrony had higher pain thresholds and felt closer than those dancing out of sync.

Pain threshold is a stand-in for endorphins, and most of these studies were small, so read them as a strong hint rather than proof. You can feel the effect at home with a round: one tune, two singers, the second starting two bars later so the lines fit together.

An original round: Voice 2 starts when Voice 1 reaches bar 3, and a third voice can join at bar 5. Every note belongs to the C chord or, at the start of each even-numbered bar, the G7 chord, so the lines sound together without clashing.

Stress and anxiety: what the meta-analyses show

Music's reputation for calming people rests on more than a hundred randomized trials. The broadest summary, a 2020 meta-analysis by Martina de Witte and colleagues, pooled 104 randomized trials with 9,617 participants: music interventions lowered stress measured in the body (heart rate, blood pressure and hormones, effect size d = 0.38) and in how stressed people felt (d = 0.55). That is a small-to-medium effect: real and worth having, not a cure.

For anxiety, a 2021 meta-analysis of 32 randomized trials of music therapy (1,924 people) found less anxiety at the end of treatment but no clear effect at later follow-up. A 2013 Cochrane review of 26 trials found that people who listened to recorded music before surgery felt less anxious than those given standard care, with a catch the authors flag: you cannot hide music from the person hearing it, so most trials carry a high risk of bias.

Cortisol, the "stress hormone", is where popular claims run ahead of the data. In a 2013 experiment, 60 women heard relaxing choral music (Allegri's Miserere), rippling water or silence before a standard stress test. The music group's cortisol response was the highest of the three, though their nervous systems recovered faster afterward. In daily life, when 55 students logged their listening and stress four times a day, listening went with lower stress, and listening to relax went with lower cortisol in the 25 who gave saliva samples. A 2018 review of 44 studies found a cortisol-lowering effect in half of the clinical studies. Music can lower stress, but context matters, and no single hormone reading proves it.

Depression: what Cochrane and later reviews found

For depression, the strongest evidence concerns music therapy delivered by trained therapists, not playlists. The 2017 Cochrane review, the most recent version, found nine studies with 421 people. Adding music therapy to usual care reduced depressive symptoms more than usual care alone in the short term, and the effect was large (a standardized mean difference of about −0.9, on both clinician-rated and self-rated scales), with moderate-certainty evidence. It also eased anxiety and improved everyday functioning, but not quality of life. The limits matter: each pooled result rests on only four small studies (about 140 to 220 people), only short-term effects were measured, and whether music therapy beats psychological therapy remains uncertain.

A 2025 meta-analysis in BJPsych Open pooled 25 randomized trials published from 2014 to 2023 and again found fewer depressive symptoms with music therapy (a large average effect, a standardized mean difference of −0.97), but only a non-significant trend for anxiety, and rated the certainty of the evidence very low because of bias, inconsistent results and imprecision. Its authors concluded that music therapy "may be recommended with weak strength": promising, not yet proven to the standard of the best treatments.

Group singing has a large recent trial. In SHAPER-PND, published in 2025, 199 mothers with symptoms of postnatal depression were randomized to ten weeks of group singing or to existing mother-and-baby activities. Both groups improved by week 10, but the singers kept slightly lower scores on a 30-point depression scale at weeks 20 and 36 (about 1.5 points) and were more likely to stay in the study. It was one unblinded trial, and the uneven dropout could itself have tilted the results.

Playing an instrument is a different question, and a large twin study shows why correlation misleads. Among 10,776 Swedish twins, people who played an instrument reported slightly more mental health symptoms, but the link weakened when twins were compared within the same family, pointing to shared genes or upbringing rather than to music. Neither "music makes you depressed" nor "learning an instrument prevents depression" follows from who happens to play.

Dementia: what music can and cannot do

The 2025 update of the Cochrane review on dementia (30 studies, 1,720 people, most in residential care) notes that responsiveness to music may remain into the late stages. At least five sessions of music-based therapy probably improved depressive symptoms slightly at the end of treatment and may have reduced behavior problems overall, but likely did not reduce agitation or aggression and showed no clear benefit for anxiety, well-being, or memory and thinking. No effect was shown four weeks or more after treatment ended. Personal playlists are worth offering for comfort, but, as the American Music Therapy Association points out, a headphone program such as Music & Memory does not meet the definition of music therapy.

The evidence at a glance, from the reviews cited in this guide. Certainty labels are each review's own rating where it gave one; the stress and anxiety meta-analyses did not grade theirs.

What a board-certified music therapist does

Music therapy is a profession, not a playlist. The American Music Therapy Association defines it as "the clinical & evidence-based use of music interventions to accomplish individualized goals within a therapeutic relationship by a credentialed professional who has completed an approved music therapy program." In practice a music therapist assesses how you respond to music, sets goals with you and your care team, and works toward them through improvisation, singing and playing, songwriting, lyric discussion, guided listening and imagery. You do not need any musical skill.

A patient of the National Intrepid Center of Excellence performs with an Air Force Band keyboard player at the monthly Creative Arts Café at Walter Reed, which began as a collaboration between the band and the center's music therapy program, December 2016. Photo: Airman 1st Class Rustie Kramer, U.S. Air Force, public domain, via Wikimedia Commons.

In the US, look for MT-BC (Music Therapist–Board Certified). It takes a bachelor's degree or higher from an AMTA-approved program, 1,200 hours of clinical training including an internship, and a national exam from the Certification Board for Music Therapists; the credential is valid for five years. In the UK, "music therapist" is a protected title on the Health and Care Professions Council register, and in Australia the Australian Music Therapy Association lists registered music therapists.

Four ways music is used for well-being. Only music therapy involves a credentialed therapist, personal goals and a therapeutic relationship.

Researchers separate music therapy from "music medicine", recorded music offered by health workers as part of care. A 2022 meta-analysis of 47 studies found a medium-to-large average effect of music therapy on stress (d = 0.72), though effects were larger in less rigorous designs and with waiting-list controls.

The work of a music therapist in four steps, summarized from the American Music Therapy Association's description.

Listening, singing and playing: what each adds

Each way of taking part asks something different, and the evidence differs too. A systematic review of music and well-being in adults found "reliable evidence for positive effects", and the activities studied most often were listening and regular group singing. Listening asks the least of you; singing with others adds synchrony and bonding. Playing and learning add something studies measure less well: the steady experience of getting better at something hard.

That mastery is real, but the evidence for its mental-health effects is thin. In a small Barcelona study, 13 older adults who took piano lessons and practiced daily for four months reported less depression and better quality of life afterward than 16 peers in other leisure activities. What learning can promise is a visible record of progress, and our How to practice sheet builds one in: a weekly log where you note the hard spot you fixed each day. For the brain science of playing, see How Learning an Instrument Rewires Your Brain.

Improvising and performing are less studied still. One small study followed 15 professional choir singers through the same program on two evenings. Sung without an audience, their levels of the stress hormone cortisol fell; sung at a live concert, they rose, and the singers' anxiety was highest just before the concert and fell steeply afterward. Sharing what you have learned brings its own stress, so starting small, with one friend, keeps the stakes low.

A 10-minute music routine for calm

You can do this at any piano or keyboard. It rests on two findings: slow breathing with long exhales tends to calm the body, and slow, soft, smooth music in a major key is what listeners hear as peaceful.

1. Breathe with the melody (minutes 0–2). Set a metronome to 60, as on Tempo & the Metronome. Play the four rising notes as you breathe in, then hum or play the falling notes as you breathe out for six counts. One breath takes ten seconds: six breaths a minute.
2. Improvise on the black keys (2–5). Hold the damper pedal down, rest your left hand on G♭ and the D♭ above it, and let your right hand wander slowly over the black keys.
3. Play or sing something slow (5–8). A gentle piece you know, or the round with a friend.
4. Listen (8–10). One calm track, eyes closed.

Slow breathing (under ten breaths a minute) goes with higher heart rate variability and feelings of calm in a 2018 systematic review, and in a 2023 randomized study, five minutes a day of breathing with long exhales improved mood more than mindfulness meditation. The melody below sets that breath to music.

Step 1: four counts in, six counts out at quarter note = 60. The bar line marks the turn of the breath; repeat the pair six times for one minute.

For step 2, find the black keys on The piano keyboard: they come in groups of two and three. Together they form a pentatonic scale with no half steps, so no two notes clash sharply; our Pentatonic & Blues Scales sheet calls this the no-fail test.

The five black keys, G♭ A♭ B♭ D♭ E♭, repeat in every octave and form the G♭ major pentatonic scale.

Start from the four bars below, then make up your own question and answer as Improvising with the Pentatonic shows. The Damper Pedal explains the pedal if it is new to you.

Step 2: an original starter on the black keys. The left hand holds an open fifth, G♭ and D♭, under one long pedal; the right hand asks a question that stops on E♭ in bar 2 and answers by coming home to G♭ in bar 4.

For step 3, choose a gentle piece such as La Candeur, the two-chord sway of Gymnopédie No. 1 or Bach's Prelude in C, and shape each phrase as our Phrasing sheet teaches: rise, taper, breathe. In step 4, if a track pulls you down, change it.

The routine at a glance. Use it alongside care, never instead of it.

When to talk to a professional, and where to find help now

Music is a good companion for everyday stress, but it is not enough on its own when low mood or worry starts to run your days. The US National Institute of Mental Health advises talking to a health care provider when signs of depression persist or don't go away; depression is diagnosed when symptoms such as low mood or loss of interest last most of the day, nearly every day, for at least two weeks. Anxiety that interferes with work, school or relationships deserves the same attention.

Signs that it is time to talk to a doctor or therapist, based on the US National Institute of Mental Health's guidance.

If you are in crisis or thinking about suicide, talk to a person today:

• United States: call or text 988, or chat at 988lifeline.org; in a life-threatening emergency, call 911.
• Canada: call or text 9-8-8; if your safety is at risk, call 911.
• UK and Ireland: call Samaritans on 116 123, free, day or night; in an emergency, call 999 in the UK, or 112 or 999 in Ireland.
• Australia: call Lifeline on 13 11 14 or text 0477 13 11 14; if life is in danger, call 000.
• Anywhere else: findahelpline.com lists free, confidential helplines in more than 175 countries.

Free worksheets for this lesson

• Major & minor: two colors of home — one lowered third turns C major into C minor
• Parallel vs Relative Minor — C major, A minor and C minor side by side
• Cadences — the authentic, plagal and half cadences, with audio
• Cadences in Every Key — perfect, plagal, half and deceptive cadences in any key
• Ear Training: Cadences & Progressions — name each cadence by ear
• Non-chord tones — passing tones, suspensions, appoggiaturas and more
• Suspended chords: sus2 and sus4 — the 4–3 fall that makes a suspension sigh
• Inside Chopin's Prelude in E Minor — the sinking chords and the climax
• The Dynamic Ladder — six volumes from pp to ff
• Phrasing — Ode to Joy as question and answer: rise, taper, breathe
• Tempo & the Metronome — tempo words and metronome marks, with audio
• The piano keyboard — the groups of two and three black keys
• The Pentatonic & Blues Scales — the black-key pentatonic and the no-fail test
• Improvising with the Pentatonic — five safe notes and a question-and-answer recipe
• The Damper Pedal — how the pedal works and how to change it
• Inside Burgmüller's La Candeur — a singing line over a quiet left hand
• The Chords of Gymnopédie No. 1 — two chords and a slow sway
• Inside Bach's Prelude in C — one broken chord in every bar
• How to practice — five steps per session and a weekly practice log

Keep music in your week

Music won't fix everything, and the research says so plainly. What it offers is pleasure through your reward system, cues that shape your mood, a fast way to feel close to other people and modest help with stress and anxiety; and, in therapeutic programs, relief of depressive symptoms for people living with depression (and, more modestly, dementia). A Tissimo teacher can help with the musical side of the habit, from a gentle evening piece to a first improvisation; browse our music teachers whenever you're ready.

Sources

• Anatomically distinct dopamine release during anticipation and experience of peak emotion to music (Salimpoor et al., 2011) — Nature Neuroscience
• Dopamine modulates the reward experiences elicited by music (Ferreri et al., 2019) — PNAS, via PMC
• Emotional expression in music: contribution, linearity, and additivity of primary musical cues (Eerola et al., 2013) — Frontiers in Psychology, via PMC
• Group music performance causes elevated pain thresholds and social bonding in small and large groups of singers (Weinstein et al., 2016) — Evolution and Human Behavior, via PMC
• Effects of music interventions on stress-related outcomes: a systematic review and two meta-analyses (de Witte et al., 2020) — Health Psychology Review
• The effect of music on the human stress response (Thoma et al., 2013) — PLOS ONE, via PMC
• Music therapy for depression (Aalbers et al., 2017) — Cochrane Database of Systematic Reviews, via PMC
• Music therapy for patients with depression: systematic review and meta-analysis of randomised controlled trials (Lee et al., 2025) — BJPsych Open, via PMC
• A community singing intervention for postnatal depressive symptoms, SHAPER-PND (Bind et al., 2025) — British Journal of Psychiatry, via PMC
• Music-based therapeutic interventions for people with dementia (van der Steen et al., 2025) — Cochrane Database of Systematic Reviews, via PMC
• Low-stress and high-stress singing have contrasting effects on glucocorticoid response (Fancourt et al., 2015) — Frontiers in Psychology, via PMC
• What is Music Therapy? — American Music Therapy Association
• 988 Suicide & Crisis Lifeline