Melodies of the Mind: How Music Shapes Your Brain and Boosts Mental Health

Global Watchdog is reader-supported, with no advertisers and no government entity to answer to. If you enjoyed this article, you can help keep it coming with a one-time donation or monthly support: Support on PatreonDonate on Ko-fi

22 min read

Last updated November 2, 2026

Last updated November 2, 2026: Full sourcing pass and refresh. Every direct quote in the 2024 version was removed after checking showed the papers cited for them do not contain them, mismatched citations were corrected or cut (the dopamine paper below is Salimpoor 2011, not 2013, and it measured something more specific than the old text claimed), the repeated opening was merged, and each claim was regraded against the reviews that actually decide it, including the 2025 Cochrane update on music and dementia, the Sala and Gobet meta-analyses on music training, and the Terry meta-analysis on music and exercise. New FAQ, reader poll, comparison table, and Bottom Line added.

A quick note before we start: this article is general information about music, mood, and the brain. It is not medical advice. Music therapy in the studies below means work with a qualified music therapist, and it was tested as an addition to regular care, not a replacement for it. If you are dealing with depression, anxiety, or dementia care, please work with your own clinicians.

A song you have not heard in thirty years comes on, and for three minutes you are seventeen again, in a particular car, with particular people, smelling a particular summer. That is not sentimentality. That is your brain doing something measurable, and scientists have spent decades mapping it with scanners, blood flow tracers, and, in one famous case, a radioactive tracer that lights up where dopamine is released.

Here is the honest shape of what they found, because the 2024 version of this article oversold parts of it and miscited the rest. Music really does engage the brain’s reward system. It really can lift mood, ease perceived effort in a workout, and sometimes unlock memories in people with dementia when other doors have closed. What it does not do is raise general intelligence, rewire a child’s brain into a better student’s brain, or treat depression and dementia by itself. The real story is smaller than the headlines and, to my mind, more interesting.

Professor Winston Wants to Know

Our resident watchdog is taking the temperature of the room.

What does music do most reliably for you?

One tap, no account or email needed. Results are shown after you vote.

What does music do most reliably for you?

Lifts my mood when little else doesBrings back people and places from my pastMakes exercise or work feel easierHelps me focus or fall asleep

Cast your voteSee results

Want to say more? The comments below are open.

How Music Reaches So Much of the Brain at Once

Most sounds ask one part of the brain to do one job. Music is rude about that division of labor. Following a melody means tracking pitch and timing in the auditory cortex, predicting what comes next, tapping a foot or fighting the urge to (motor areas), feeling something about it (limbic structures including the amygdala), and, if the song is an old friend, pulling up the memory of where you were when it mattered (the hippocampus and medial prefrontal areas). Reviews of the cognitive neuroscience of music, including Toader and colleagues (2023) in Brain Sciences, describe exactly this spread: music is one of the few everyday activities that recruits perception, movement, emotion, and memory in a single pass. Their review is here: Toader et al. 2023, Cognitive Crescendo.

That breadth is why music researchers get excited, and also why sloppy claims breed so easily. A stimulus that touches everything can be credited with improving anything, and a citation can be stapled to almost any sentence and look plausible. Several of the 2024 citations in this article were stapled exactly that way, to papers about film music, orchestral musicians’ vision, and happy and sad faces in a scanner. They are gone. What follows cites each paper for what it actually did.

Why a Favorite Song Can Hit Like a Reward

The chills study

Blood and Zatorre (2001) scanned ten McGill students, all trained musicians chosen because specific pieces reliably gave them chills, while they listened to their own selections (Rachmaninoff and Barber were on the program) and to other people’s. Using PET blood flow imaging, they found that as chills intensified, activity rose and fell in the ventral striatum, midbrain, amygdala, orbitofrontal cortex, and ventral medial prefrontal cortex, a circuit the authors noted is also recruited by food, sex, and drugs of abuse. Read the study record here: Blood and Zatorre on PubMed. Two limits belong in the same paragraph: ten people is a tiny sample, and blood flow is not dopamine. This study shows where the response lives, not the chemistry of it.

The dopamine study, dated correctly

The chemistry came a decade later. Salimpoor and colleagues (2011) used raclopride PET, in which reduced tracer binding is read as endogenous dopamine release, in listeners selected for reliably getting chills. Dopamine rose in the striatum at peak emotional moments, with a neat split in the companion imaging: the caudate was more involved during anticipation, the run-up to the peak, and the nucleus accumbens during the peak itself. That anticipation finding is the one I keep thinking about. Your brain does not just enjoy the chorus. It savors the verse because it knows the chorus is coming. Study record: Salimpoor et al. 2011, Nature Neuroscience.

New music, same reward system

What about songs you have never heard? Salimpoor and colleagues (2013), this time in Science and using fMRI rather than any dopamine measure, played 60 unheard excerpts while participants bid real money for them in an auction. Nucleus accumbens activity during listening was the best predictor of what a song was worth to that listener, and the value signal depended on how strongly the auditory and frontal regions were talking to the accumbens. In plain terms, new music pays off when your pattern-matching cortex convinces your reward system the pattern is worth keeping. Study record: Salimpoor et al. 2013 on PubMed.

Why Familiar Songs Feel Bigger Than New Ones

Part of the answer is the mere exposure effect: repeated listening generally increases liking, up to the point where repetition curdles into irritation, a curve every radio programmer and every parent of a toddler knows by heart. The deeper part is autobiographical. Janata (2009) scanned students listening to chart songs from their own adolescence and found that familiar songs that triggered personal memories engaged the dorsal medial prefrontal cortex, a region that sits at the crossroads of music, emotion, and self-referential memory. The songs were not just recognized. They were filed under my life. Study record: Janata 2009 on PubMed.

That filing system explains the experience at the top of this article, and it also explains the dementia findings below. A song learned at seventeen is not stored in one place. It is braided into the people, rooms, and feelings of being seventeen, and braids hold when single threads snap.

Music Training: Good for Musicians, Not a Shortcut to a Smarter Brain

Start with the study that launched a thousand lesson brochures. Chan, Ho, and Cheung (1998), in a one-page report in Nature, compared 60 female university students in Hong Kong: 30 with at least six years of instrumental training begun before age twelve, 30 with none. The trained group recalled significantly more words from a spoken list. Visual memory did not differ. It is a real result, and it is also cross-sectional, which means it cannot show the training caused the difference. Good verbal learners may simply be more likely to start lessons, enjoy them, and stay. Study record: Chan et al. 1998, Nature.

The best experimental evidence is more modest and more interesting. Schellenberg (2004) randomly assigned 144 six-year-olds to a year of keyboard lessons, voice lessons, drama lessons, or no lessons. Every group gained IQ points over the year, as six-year-olds do. The music groups gained a little more, roughly 2.7 points net on full-scale IQ, an effect the author himself described as relatively small. Study record: Schellenberg 2004 on PubMed.

Then the meta-analyses grew up. Sala and Gobet (2017) pooled the literature and found a small overall effect that shrank as study quality rose. Their 2020 multilevel meta-analysis in Memory and Cognition covered 54 studies and 6,984 children: the small effects appeared only in studies without random allocation and without active control groups, and once design quality was controlled, the overall effect on cognitive and academic outcomes was essentially null, for every outcome type, age, and training length they tested. Study record: Sala and Gobet 2020.

My ruling as a retired teacher: buy your child music lessons because music is worth knowing, because discipline and ensemble playing and a room full of people breathing together before a downbeat are worth knowing. Do not buy them as an IQ appliance. A purchase made on a false promise gets returned, and the music gets blamed.

The Mozart Effect, Retired

The original finding, before the baby-product industry got to it: Rauscher, Shaw, and Ky (1993) gave 36 college students ten minutes of Mozart’s Sonata for Two Pianos in D major, K.448, relaxation instructions, or silence, then tested spatial reasoning. The Mozart group scored the equivalent of 8 to 9 points higher on the spatial subtask scores. Not on general intelligence, which was never measured. The effect lasted no more than about 10 to 15 minutes. Study record: Rauscher et al. 1993 on PubMed.

Pietschnig, Voracek, and Formann (2010) pooled nearly 40 studies and more than 3,000 participants. Mozart versus silence or a non-musical stimulus showed a small effect (d = 0.37). Other music versus the same controls showed essentially the same effect (d = 0.38). Mozart versus other music was negligible (d = 0.15), and the analysis found publication bias pushing the numbers up, with larger effects in studies from labs affiliated with the original researchers. The parsimonious explanation, which Schellenberg’s own listening studies support, is enjoyment arousal: music you like wakes you up and puts you in a better mood, and an awake, cheerful brain does a spatial puzzle a little better for a few minutes. A story you enjoy can do the same. Study record: Pietschnig et al. 2010.

So no, playing Mozart to a baby does not build a smarter baby, and it never did. Playing music to a baby because babies like music and parents like sharing it needs no help from neuroscience.

Music, Memory, and Dementia: A Door That Sometimes Still Opens

Here the evidence is genuinely moving, and it still has to be graded carefully, because families caring for someone with dementia are exactly the readers who most deserve not to be sold magic.

Why long-known music survives in the brain

Jacobsen and colleagues (2015), in Brain, first mapped where long-known music is encoded in healthy young adults, landing on the caudal anterior cingulate cortex and the ventral pre-supplementary motor area. Then they looked at those regions in Alzheimer’s patients and found substantially minimal cortical atrophy and minimal disruption of glucose metabolism there compared with the rest of the brain. One honest caveat the authors themselves build in: amyloid deposition in those regions was not lower, so the regions look early-stage rather than untouched, and the Alzheimer’s group was not directly tested on musical memory performance. It is a sparing explanation, not proof of a cure mechanism. Study record: Jacobsen et al. 2015 on PubMed.

Songs still pull up a life

Baird and colleagues (2018) played famous songs and showed famous-event photographs to 10 people with Alzheimer’s disease and 10 healthy older adults. In the healthy group, photographs evoked more autobiographical memories than songs. In the Alzheimer’s group, that gap disappeared: song-evoked memories occurred at a frequency that did not differ from the healthy controls, while photograph-evoked memories had fallen away. The song-evoked memories were typically less specific, more like general knowledge of an era than a single detailed afternoon. Songs from ages roughly 10 to 30, the reminiscence bump years, worked best. Study record: Baird et al. 2018 on PubMed. And a caution from Baird and Samson’s earlier review: musical memory is not one thing, and not all of it is preserved. Recognition of a beloved song and the ability to hum along can survive while memory for new melodies is plainly impaired.

If you want the fuller picture of what memory science supports and what it does not, our companion guide, keep your memory sharp: what actually works, grades the whole field the same way.

What the Cochrane reviewers will and will not sign

The Cochrane review of music-based interventions for dementia was updated in 2025 (van der Steen et al.; 30 studies, 1,720 participants). Against usual care, the finding that survived is a small reduction in depressive symptoms (SMD -0.23, moderate certainty). Cognition did not improve. Agitation and aggression likely did not improve. Emotional well-being and anxiety, which the 2018 version thought might improve, did not improve in the updated analysis, and there was no evidence of long-term effects after the sessions ended. Review record: van der Steen et al. 2025 on PubMed.

Put together, the honest pitch for a dementia playlist is this: it will probably not change the disease, the test scores, or the hard afternoons in general. It may lift mood a little during and after listening, and it may hand you back a person for the length of a song, present and themselves, humming. Anyone who has sat in that room knows that is not a small thing. It is just not medicine, and the distinction matters because medicine is what families get billed for.

Music Therapy for Depression, Anxiety, and Stress

Definitions first, because the marketplace blurs them on purpose. Music therapy is a clinical practice: regular sessions with a qualified music therapist, making and listening to music as treatment. Putting on a playlist you love is a music intervention, which is a broader and cheaper thing. Both have evidence. They are not the same evidence.

For depression, the Cochrane review by Aalbers and colleagues (2017) pooled 9 studies and 421 participants and found that music therapy added to treatment as usual beat treatment as usual alone in the short term, with large effects on clinician-rated depressive symptoms (SMD -0.98) and patient-reported symptoms (SMD -0.85), rated moderate certainty. The same review is candid about the edges: the confidence intervals are wide, quality of life did not improve, and whether music therapy matches psychological therapy head to head is uncertain on mostly low-quality evidence. Review record: Aalbers et al. 2017 on PubMed.

For stress, de Witte and colleagues (2020) meta-analyzed 104 randomized trials and 9,617 participants across music interventions and found a medium psychological stress reduction effect (d = 0.545) and a smaller physiological one (d = 0.380, including heart rate, blood pressure, and hormone measures). Their companion analysis of therapist-delivered music therapy alone found a larger pooled effect (d = 0.723) but flagged the pattern that should keep you honest: effects were bigger against waitlist controls than against active care, and bigger in non-randomized designs, which is what inflated estimates look like. Review record: de Witte et al. 2020 on PubMed.

So: music as an adjunct for mood and stress has real, replicated support, strongest in the short term and alongside regular care. Music instead of care has none. If a product, app, or practitioner tells you otherwise, you have learned something about the seller, not about music.

Music and Exercise: The Strongest Everyday Case

If you want the music effect with the least argument around it, put on your shoes. Terry, Karageorghis, and colleagues (2020) meta-analyzed 139 studies, 598 effect sizes, and 3,599 participants in Psychological Bulletin. Music’s largest effect was on how exercise feels (affective valence, g = 0.48). Performance improved moderately (g = 0.31). Perceived exertion dropped a little (g = 0.22), oxygen consumption improved a little (g = 0.15), and heart rate did not change significantly. Effects were larger for exercise than for sport, and larger for fast tempo music. Review record: Terry et al. 2020.

Karageorghis’s research program supplies the mechanism and the limit. Music works mainly by dissociation at low to moderate intensity: it competes with fatigue signals for your attention, improves your mood, and lets the miles feel shorter. At very high intensity, the body’s signals win the argument no matter what is playing, which every runner who has ever hit the last mile already knows. The famous claim that music boosts endurance by 15 percent traces to small treadmill studies with synchronous motivational music, a few dozen participants at most. The pooled effect above is the number to trust, and it is plenty: a workout that feels better is a workout that happens again tomorrow. For tracking what those workouts actually do, see our Apple Health app review.

One safety line that outranks every finding in this section: if you exercise on roads, keep one ear open or the volume low enough to hear traffic. No playlist is worth a blind intersection.

How to Use Music on Purpose, Without Believing Magic

• For mood: keep a short playlist of songs with a proven personal track record, the ones that have actually lifted you before, and use it early in a slump rather than at the bottom of one. Familiar beats novel here, because familiar is where the autobiographical wiring lives.
• For exercise: pick fast tempo, self-selected music you genuinely like, and save the best tracks for the segment you usually quit in. Expect to feel better first and perform a little better second.
• For a loved one with dementia: build the playlist from their teens and twenties, not yours. Sit with them, watch the reaction, and stop any song that agitates rather than settles. Treat it as a visit, not a treatment session, and tell the care team what worked.
• For focus: the honest evidence is mostly about arousal and mood, so run the experiment on yourself. Lyrics compete with words if your task is verbal. Familiar, low-drama instrumental music is the usual winner, and silence is a legitimate control condition.
• For sleep: slow, familiar music as part of a wind-down routine is a reasonable personal experiment, not a sleep therapy. If broken sleep is a pattern rather than a bad week, our science of sleeping in article is a useful next read, and persistent insomnia deserves a clinician, not a longer playlist.

Watchdog Approved

Winston’s Pick: Winston hears frequencies you cannot. Trust his taste in listening gear.

Sony WH-1000XM5 Noise-Canceling Headphones (Midnight Blue)
Industry-leading noise cancellation for losing yourself in the music, or in your focus playlist.

Check Price on Amazon

Apple AirPods Pro, 2nd Generation (USB-C)
The everyday earbuds for music on the move, with surprisingly good noise cancellation.

Check Price on Amazon

Audio-Technica AT-LP60X-BK Turntable
A beloved starter turntable for rediscovering vinyl: warm sound, zero fuss.

Check Price on Amazon

As an Amazon Associate, we earn from qualifying purchases.

Frequently Asked Questions

Does listening to Mozart make you or your baby smarter?

No. The 1993 study behind the claim tested 36 college students on one spatial subtask, found a gain worth 8 to 9 points on that subtask’s scoring, and watched it fade within about 15 minutes. General intelligence was never measured. In the largest meta-analysis, other music performed the same as Mozart, which points to arousal and mood, not a special property of one composer. Enjoy Mozart for Mozart.

Do music lessons raise a child’s IQ or grades?

Not reliably, on current evidence. One well-known randomized trial found a small IQ advantage of about 2.7 points after a year of lessons in six-year-olds (Schellenberg, 2004). But the 2020 multilevel meta-analysis of 54 studies and 6,984 children found that once random assignment and active control groups are required, the effect on cognitive and academic outcomes is essentially zero (Sala and Gobet, 2020). Lessons reliably produce musical skill, which is reason enough.

Is music therapy the same thing as listening to my favorite playlist?

No. Music therapy is treatment delivered by a qualified music therapist, in regular sessions, usually added to standard care. That is what the Cochrane depression review tested. A personal playlist is a music intervention: cheaper, pleasant, and supported for short-term stress relief, but it has not been shown to treat depression. Use playlists freely. Just do not let anyone bill one as therapy.

Can music bring back memories for someone with Alzheimer’s disease?

Sometimes, in a specific and limited way. Familiar songs evoked autobiographical memories in Alzheimer’s patients at a frequency similar to healthy older adults in Baird et al. (2018), even as photograph-evoked memories declined, and the brain regions encoding long-known music show relatively little atrophy (Jacobsen et al., 2015). The memories tend to be general rather than detailed, music does not restore memory or improve cognition scores in the Cochrane data, and a song that once comforted can occasionally distress. Build the playlist from their youth, stay for the reaction, and let their face grade the intervention.

What music is best for workouts?

Fast tempo music you chose yourself and actually like. In Terry et al. (2020), music’s biggest exercise effect was on how the effort felt (g = 0.48), with a moderate performance gain (g = 0.31) and a small drop in perceived exertion (g = 0.22). Self-selection matters because preference is part of the mechanism. Keep the volume safe and one ear on traffic if you are near roads.

Sources

• Blood, A. J., and Zatorre, R. J. (2001). Intensely pleasurable responses to music correlate with activity in brain regions implicated in reward and emotion. Proceedings of the National Academy of Sciences, 98(20), 11818-11823. PubMed 11573015
• Salimpoor, V. N., Benovoy, M., Larcher, K., Dagher, A., and Zatorre, R. J. (2011). Anatomically distinct dopamine release during anticipation and experience of peak emotion to music. Nature Neuroscience, 14(2), 257-262. doi:10.1038/nn.2726
• Salimpoor, V. N., van den Bosch, I., Kovacevic, N., McIntosh, A. R., Dagher, A., and Zatorre, R. J. (2013). Interactions between the nucleus accumbens and auditory cortices predict music reward value. Science, 340(6129), 216-219. PubMed 23580531
• Zatorre, R. J., and Salimpoor, V. N. (2013). From perception to pleasure: Music and its neural substrates. Proceedings of the National Academy of Sciences, 110 (Suppl 2), 10430-10437. Background review for the reward model.
• Janata, P. (2009). The neural architecture of music-evoked autobiographical memories. Cerebral Cortex, 19(11), 2579-2594. PubMed 19240137
• Toader, C., Tataru, C. P., Florian, I. A., et al. (2023). Cognitive Crescendo: How Music Shapes the Brain’s Structure and Function. Brain Sciences, 13(10), 1390. PMC10605363. Background review, cited only for the breadth of brain systems music engages. Note: the 2024 version of this article attributed this review to Bigand and Tillmann (2018); no such paper exists in the record, and the attribution was a fabrication in the original reference list.
• Chan, A. S., Ho, Y. C., and Cheung, M. C. (1998). Music training improves verbal memory. Nature, 396(6707), 128. doi:10.1038/24075
• Schellenberg, E. G. (2004). Music lessons enhance IQ. Psychological Science, 15(8), 511-514. PubMed 15270994
• Sala, G., and Gobet, F. (2020). Cognitive and academic benefits of music training with children: A multilevel meta-analysis. Memory and Cognition, 48(8), 1429-1441. doi:10.3758/s13421-020-01060-2; and Sala and Gobet (2017), Educational Research Review, 20, 55-67.
• Rauscher, F. H., Shaw, G. L., and Ky, C. N. (1993). Music and spatial task performance. Nature, 365(6447), 611. PubMed 8413624
• Pietschnig, J., Voracek, M., and Formann, A. K. (2010). Mozart effect-Shmozart effect: A meta-analysis. Intelligence, 38(3), 314-323. doi:10.1016/j.intell.2010.03.001
• Jacobsen, J. H., Stelzer, J., Fritz, T. H., Chetelat, G., La Joie, R., and Turner, R. (2015). Why musical memory can be preserved in advanced Alzheimer’s disease. Brain, 138(Pt 8), 2438-2450. PubMed 26041611
• Baird, A., Brancatisano, O., Gelding, R., and Thompson, W. F. (2018). Characterization of music and photograph evoked autobiographical memories in people with Alzheimer’s disease. Journal of Alzheimer’s Disease, 66(2), 693-706. PubMed 30320586; and Baird and Samson (2009), Neuropsychology Review, 19(1), 85-101.
• Aalbers, S., et al. (2017). Music therapy for depression. Cochrane Database of Systematic Reviews, 11, CD004517. PubMed 29144545
• de Witte, M., Spruit, A., van Hooren, S., Moonen, X., and Stams, G. J. (2020). Effects of music interventions on stress-related outcomes: a systematic review and two meta-analyses. Health Psychology Review, 14(2), 294-324. PubMed 31167611
• van der Steen, J. T., et al. (2025). Music-based therapeutic interventions for people with dementia. Cochrane Database of Systematic Reviews, CD003477 (update; evidence current to November 30, 2023). PubMed 40049590
• Terry, P. C., Karageorghis, C. I., Curran, M. L., Martin, O. V., and Parsons-Smith, R. L. (2020). Effects of music in exercise and sport: A meta-analytic review. Psychological Bulletin, 146(2), 91-117. doi:10.1037/bul0000216

As an Amazon Associate, we earn from qualifying purchases.

So tell me: what is the song that stops you in your tracks, the one that puts a whole room, a whole year, a whole person back in front of you before the first chorus ends? Name it, and tell me where it takes you. I read every answer, and I suspect the pattern in yours will say more about how music shapes your brain than any scanner either of us will ever lie in.