Music helped the brain "turn off" stress faster: scientists traced the pathway from the auditory cortex to the insula and thalamus.

Music can indeed help the body recover more quickly from acute stress—and a new experiment suggests it's likely not just about pleasant emotions or distraction. Scientists from the Institute of Psychology of the Chinese Academy of Sciences discovered that listening to music after a stressful task restructures the interactions between the auditory cortex, insular cortex, and thalamus—the brain regions that link sound processing with the perception of one's own state and the regulation of the stress response. The study was published in the Proceedings of the National Academy of Sciences. [1]

A total of 120 healthy volunteers participated in two experiments. To create controlled acute stress, the researchers used a time-limited mental arithmetic task. Afterward, some participants listened to relaxing music, while the control group listened to calming natural sounds, such as running water. The researchers simultaneously monitored mood, heart rate, electrodermal activity (EDA), heart rate variability, cortisol, and, in a second experimental series, functional brain activity using magnetic resonance imaging. [2]

Both types of sound were associated with a gradual, natural recovery from stress, but music provided an additional benefit. Those listening to music experienced a faster reduction in heart rate and electrodermal activity, a more pronounced recovery in positive mood, and a more favorable cortisol profile was also observed in a functional magnetic resonance imaging experiment. Importantly, listening to music before the stressor did not prevent the stress response itself. The most pronounced effect emerged precisely after the stressor ceased. [3]

Even more intriguing was the mechanism. Music didn't simply activate or suppress a single brain region. Dynamic causal modeling and time-varying functional connectivity analysis showed that it restructured information transmission within the auditory cortex → insular cortex → thalamus network. Moreover, the dynamics of this restructuring were linked to musical tension—the sensation of anticipation, buildup, and resolution that arises as the melody and harmony develop. This allowed the authors to propose a more complex model: the structured progression of music may serve as a kind of temporary signal, helping the nervous system return the stress response to its original state. [4]

Key parameter	Research data
Participants	120 healthy volunteers
Number of experiments	2
Primary stressor	time-limited mental arithmetic
Comparison	relaxing music vs. soothing natural sounds
Size of experimental subgroups	about 30 people each
Psychological indicators	positive and negative mood
Autonomous indicators	Heart rate, electrodermal activity, HRV
Endocrine indicator	salivary cortisol
Neuroimaging	functional MRI
Brain connection analysis	dynamic causal modeling + dynamic functional connectivity
Main network	auditory cortex → insula → thalamus
The main effect	accelerating recovery from, rather than preventing, stress
DOI	10.1073/pnas.2611584123

Why is it important to distinguish between the stress response and recovery from it?

Stress itself is not a pathology. When a person needs to quickly solve a complex problem, speak in front of an audience, or respond to a potential threat, the sympathetic nervous system accelerates the heart rate, alters sweating, and mobilizes energy reserves. Simultaneously, the hypothalamic-pituitary-adrenal axis is activated, resulting in increased cortisol levels.

This response is beneficial as long as it corresponds to the actual stress. The problem arises when the stressor has already disappeared, but the physiological system remains in a state of heightened alert. The heart continues to beat faster, sympathetic tone remains elevated, emotional tension persists, and the brain continues to behave for some time as if the threat is not over.

This is why the authors of the new study examined stress recovery as a separate process. Most common advice about relaxing music suggests that it simply "keeps us from getting nervous." However, the experiment revealed a more interesting picture: listening to music beforehand did not significantly diminish the stress response to the subsequent arithmetic task. [5]

The benefit emerged after the stressful task had already ended. Compared to natural sounds, music helped several body systems return to a calm state more quickly. Therefore, the study suggests increased recovery efficiency rather than the creation of a "protective bubble" against stressful events. [6]

What the researchers measured

System	Indicator	Which roughly reflects
Emotional	positive mood	subjective recovery
Emotional	negative mood	residual emotional stress
Cardiovascular	heart rate	level of autonomic arousal
Vegetative	electrodermal activity	sympathetic activation
Vegetative	heart rate variability	dynamics of autonomic regulation
Endocrine	cortisol	activity of the stress hormonal system
Nervous	functional MRI	changes in brain activity and connectivity

The researchers compared music not to silence, but to truly relaxing sounds.

This is one of the most powerful design features. If participants had listened to music after stress, while the control group remained in complete silence, the positive result could have been explained by almost anything: the presence of a pleasant sound, distraction, or simply the nature of auditory stimulation.

Instead, the control group was exposed to calming natural sounds, including water and rain. These sounds are naturally perceived as relaxing and can accompany a natural reduction in physiological arousal. Thus, the researchers were trying to determine whether music had any additional properties compared to a standard, pleasant acoustic background. [7]

It was against this backdrop that music demonstrated its superiority. Both groups recovered over time, which is unsurprising: the acute laboratory stress subsided, and the body gradually returned to normal. However, the music group experienced a faster recovery of some parameters, and a greater improvement in positive mood. [8]

This significantly strengthens the study's conclusion. The observed effect cannot be simply reduced to the formula "pleasant sound is better than no sound." The authors actually compared two potentially calming acoustic interventions and found that structured music had a slightly different effect on recovery than natural background sounds.

The heart and sweat glands showed that the body actually recovered faster.

Heart rate is one of the simplest indicators of the autonomic stress response. During a strenuous arithmetic task, the heart rate accelerates as the sympathetic nervous system mobilizes the body. After completing the task, the heart gradually slows.

Participants listening to music experienced a faster return to normal than those listening to natural sounds. A similar pattern was observed for electrodermal activity—the electrical conductivity of the skin, which is highly dependent on the activity of sweat glands, which are controlled by the sympathetic nervous system. A more rapid decline in this indicator indicates a faster attenuation of physiological arousal. [9]

In the neuroimaging portion of the study, the researchers also analyzed heart rate variability. This is a more complex indicator, reflecting changes in the intervals between successive heartbeats and the interaction between sympathetic and parasympathetic control. The dynamics of music-related brain networks correlated with the recovery of both heart rate and heart rate variability. [10]

Importantly, this isn't just a subjective response like "I felt calmer." There were also changes in bodily parameters that the participant isn't able to consciously control to the same degree. This overlap between emotional, autonomic, and endocrine levels makes the study more compelling than a study based solely on mood questionnaires.

Positive mood was restored more strongly than with natural sounds

The emotional results were also interesting. After the stressful arithmetic task, mood predictably worsened. It then gradually recovered in both groups, but those listening to music experienced a stronger return of positive emotions. [11]

The authors don't interpret this as a simple replacement of an unpleasant emotion with a pleasant one. According to their model, the emotional effect is linked to how the brain simultaneously processes acoustic structure and the body's internal state. The insular cortex plays a particularly important role here, as it is involved in interoception—the perception of heartbeat, breathing, visceral sensations, and other internal signals.

Further support for this idea came from a directional connectivity analysis. The more the interaction between the right insula and thalamus changed in the music group, the more pronounced the improvement in positive mood. This links subjective emotional recovery to a specific neural network reorganization. [12]

But such a correlation cannot be interpreted as proof that strengthening one specific connection automatically "creates a good mood." It shows that both processes changed in concert. A rigorous causal test of a specific brain pathway would require directly interfering with its function, for example, through neuromodulation techniques.

Cortisol confirms that the effect was not limited to the autonomic nervous system

Cortisol is the final hormonal product of the hypothalamic-pituitary-adrenal stress axis. Unlike heart rate or skin conductance, its changes occur relatively more slowly, so it offers additional insight into recovery from a stressor.

In the second experimental series, the researchers collected saliva samples and assessed cortisol concentrations. The music group's hormonal recovery profile was found to be more favorable than that of the natural sounds group. [13]

This is important because stress is not a unified system. The sympathetic nervous system can react quickly, while the hypothalamic-pituitary-adrenal axis has its own time dynamics. When an intervention simultaneously affects mood, heart rate, sweating, and cortisol, the likelihood that only a random change in a single marker is observed becomes less likely.

However, the study does not conclude that regular music listening consistently reduces baseline cortisol levels or treats diseases associated with chronic stress. The experiment examined a single acute laboratory stressor and a relatively short recovery phase. The mechanisms may be significantly different for chronic stress.

At what levels was the effect observed?

Level	What changed in favor of music
Psychological	a more pronounced restoration of positive mood
Autonomous	heart rate decreased faster
Autonomous	skin conductivity was restored more quickly
Autonomous	the dynamics of heart rate variability changed
Endocrine	more favorable cortisol recovery
Neural	reorganization of the auditory-insular-thalamic network

MRI revealed not just one “music center,” but a restructuring of the entire stress-regulating network.

The second part of the study used functional magnetic resonance imaging. This method tracks changes in blood flow and oxygen saturation, which indirectly reflect the activity of various brain regions. The researchers were primarily interested in how the connections between regions change, rather than simply which region "lights up" more.

At the center of the discovered network was the auditory cortex, including areas of the superior temporal gyrus. This is where complex sound structure is analyzed. The changes then spread to the insular cortex—an area that integrates emotional, autonomic, and internal bodily signals—and then to the thalamus, which is involved in routing sensory information and regulating many functional states of the brain. [14]

Using dynamic causal modeling, the researchers estimated directional effective connectivity—the statistically modeled influence of one network node's activity on another. The music condition enhanced the influence of the auditory system on the insula and further altered the insula's connection to the thalamus. [15]

Terminological caution is required here. The authors of the abstract use the term "directed causal pathway," but dynamic causal modeling is a mathematical model of effective communication, not experimental stimulation of a specific neural pathway. Therefore, it is more accurate to say that the data support the "auditory cortex → insula → thalamus" model of directed influence, rather than that the study has definitively proven the physiological causality of each link.

Three main nodes of the discovered network

Region	The main role in the context of the study
Auditory cortex/superior temporal gyrus	analyzes the temporal structure of music
Insular cortex	links emotions with the state of the body and the autonomic system
Thalamus	integrates and redirects sensory and regulatory signals
Suggested direction	auditory system → insula → thalamus

The main finding was that the brain responded to musical tension over time.

Music differs fundamentally from many natural sounds in that it contains a complex structure of expectations. When a harmony or melody leads to a certain resolution, the listener unconsciously forms an expectation. If the resolution is delayed or an unexpected harmony appears, subjective musical tension arises.

This tension is not the same as psychological stress. Musical tension is an aesthetic sensation of anticipation: the melody seems to "beg for continuation" and then returns to stability. Previous research by Siqi You also showed that perceived musical tension depends on the predictability of the harmonic structure and is accompanied by measurable changes in neural processing. [16]

In a new experiment, activity in both superior temporal gyri followed changes in subjectively perceived musical tension and pleasure. Moreover, the dynamic functional connectivity between auditory areas, the insula, and the thalamus changed as musical tension increased and resolved. [17]

These network oscillations were associated with subsequent recovery of negative emotions, heart rate, and heart rate variability. Therefore, the researchers suggest that it is the temporal structure of music, rather than simply its pleasantness, that may help the nervous system gradually transition from a state of heightened alertness to a calmer configuration. [18]

Music can work through controlled alternation of anticipation and resolution

The authors' proposed model differs from traditional understandings of relaxation. It is generally believed that effective "anti-stress" music should be as monotonous, slow, and non-stimulating as possible. New data suggest that a certain degree of tension within the music may itself be functionally important.

When musical expectation is violated or delayed, the auditory system must revise its prediction. Harmonic resolution then reduces uncertainty. This sequence of tension → expectation → resolution is repeated many times within a musical work.

Researchers suggest that this structure temporarily releases the stress network from a relatively rigid state, then helps restructure the interactions of its nodes, and ultimately contributes to a reduction in overall physiological arousal. In an official report from the Institute of Psychology, this process is described as a sequential change in the interactions of the auditory system with the insula and thalamus as musical tension develops. [19]

This mechanism is currently supported by correlations between musical structure, functional connectivity, and physiological parameters. The researchers did not vary musical tension independently of all other characteristics of the piece in a full factorial experiment. Therefore, it is not yet possible to claim that musical tension is the only necessary component of the anti-stress effect.

Why ordinary water sounds were weaker than structured music

Natural sounds can also create a pleasant and predictable acoustic background. They can mask noise, reduce unexpected stimuli, and facilitate relaxation. Therefore, the control group with water and rain gradually recovered on its own.

But such sounds typically have a less hierarchical structure of expectations than music. In tonal music, the brain constantly makes predictions about the next chord, rhythmic event, or melodic resolution. Previous research shows that changing the probability of these events does indeed alter subjective stress and the brain's electrophysiological responses. [20]

The new study links this predictive mechanism to stress recovery. This is why the authors consider music not just a "pleasant background noise," but a temporarily organized signal that can actively guide the nervous system through a sequence of different states.

This doesn't mean that the sound of rain or the sea "doesn't work." They were also associated with recovery. The experiment merely shows that, in a specific laboratory situation, music produced additional changes in several parameters compared to an active natural control. [21]

It is important that music helped after stress, and did not make a person immune to it.

One of the most practical findings of the study concerns the timing of listening. If music was played before a stressful task, it did not significantly prevent the subsequent increase in stress response. [22]

This contradicts the simple notion that ten minutes of relaxing music can "protect" the body in advance from an upcoming exam, conflict, or difficult meeting. At least in the new experiment, stressful arithmetic still activated emotional and physiological systems.

However, after the stressor ended, music accelerated recovery. From a practical standpoint, this suggests that it's potentially more rational to consider it as a tool for post-stress recovery, similar to rest, rather than as a way to completely block the normal stress response.

But even this conclusion is still limited to a laboratory model. Real-life stressful situations tend to be significantly longer, and people can continue to mentally revisit the event after it's over. The study did not directly test how well the discovered mechanism works after a conflict, a workday, or prolonged emotional stress.

Is it possible to recommend a specific playlist for stress reduction?

No. The study demonstrates how music works, but it doesn't define a single optimal composition, tempo, or genre. The very fact that musical structure and subjective tension were associated with changes in brain networks suggests the need to consider the temporal dynamics of a piece rather than the existence of a universal "relaxation frequency."

Nor can one conclude that the slowest music is necessarily the best. If the proposed mechanism truly depends on the sequential emergence and resolution of expectations, then a completely tension-free, monotonous piece of music could theoretically be less effective than a composition with a gently evolving structure.

Furthermore, musical expectations depend on culture, previous experience, and familiarity with the musical system. Someone raised on Western tonal music will predict harmonic events differently than someone with a different musical tradition. Individual preferences can also influence emotional responses.

The researchers themselves cite personalized music as a logical direction for future research. It is necessary to determine whether self-selected music works better than standardized music, which characteristics are optimal for different people, and whether musical interventions can be tailored to individual physiological responses. [23]

Can music become a real therapeutic tool?

Music has clear advantages over pharmacological treatments: it's inexpensive, readily available, non-invasive, and poses virtually no medical risks at normal volumes. Therefore, even a moderate physiological effect may be of interest for stress management programs.

Potential applications include recovery from medical procedures, stressful work, exams, or other short-term stressors. Music interventions are already being studied in hospitals to reduce anxiety and pain, but the results of previous studies are mixed, in part due to differences in the music, participants, and methodology.

The new work is important because it moves beyond the question of “does music work?” to the question of “what dynamic neural system might it operate through?” Instead of a single mood indicator, the authors link the temporal structure of music to a brain network and then to autonomic and emotional recovery. [24]

But we're still a long way from a "musical recipe." For clinical use, the effect must be demonstrated in people with anxiety disorders, post-traumatic stress disorder, chronic stress, and other conditions—not just in healthy volunteers after a brief arithmetic stressor.

What the study doesn't prove

Firstly, the study does not demonstrate that music cures chronic stress. The experiment examined acute short-term stress in healthy individuals. Chronic stress is accompanied by more complex changes in sleep, the endocrine system, behavior, and sometimes mental illness.

Secondly, functional magnetic resonance imaging measures the hemodynamic signal, not the direct electrical interactions of individual neurons. Dynamic causal modeling allows one to estimate the most probable direction of connections within a predetermined network, but remains a model-based method.

Third, the observed correlation between musical tension, brain connectivity, and physiological recovery does not prove that musical tension is the sole cause of the effect. Tempo, rhythm, harmony, familiarity, subjective pleasure, and attention are closely intertwined.

Finally, 120 participants is a sufficiently large sample for a neuroimaging experiment, but too small to draw general conclusions about the human population. It's necessary to test people of different ages, cultures, musical preferences, and stress tolerance levels.

What can and cannot be concluded

The study supports	The study does not prove
Music speeds up recovery from laboratory stress	that music prevents any stress
the effect surpasses relaxing natural sounds in several respects	that natural sounds are useless
the auditory cortex-insula-thalamus network is involved	that this is the only network
musical tension is linked to network dynamics	that voltage is the only active component
heart rate, EDA, HRV, mood, and cortisol change	treatment of chronic stress diseases
DCM supports directional links	direct neural causality at the level of stimulation experiments
the result was obtained in healthy adults	the same effectiveness in clinical patients

Why this work is stronger than many previous studies on music and stress

The first strength is the use of an active control condition. Participants were compared not to silence, but to another pleasant auditory stimulus. This makes it more likely that the additional effect is due to the specific characteristics of the music.

The second was a multisystem assessment. Subjective emotions, heart rate, electrodermal activity, heart rate variability, and cortisol were simultaneously recorded. In the second series, functional magnetic resonance imaging was added. [25]

The third was an analysis of temporal dynamics. The researchers went beyond comparing average brain activity over the entire musical fragment. They analyzed how connectivity changed as musical tension built and resolved. This allowed them to link musical structure to physiological recovery on a shorter time scale. [26]

Finally, the results are consistent with previous work by first author Siqi You on the neural processing of musical expectation and tension. This group previously demonstrated that the predictability of a harmonic context alters the subjective tension curve and the brain's electrophysiological response. The new study extends the same principle of predictive musical processing to the field of stress physiology. [27]

What remains to be seen next

One obvious next step is to experimentally vary the structure of the same musical material. For example, one could create versions with the same tempo and timbre, but different degrees of harmonic tension. This would allow for a much more thorough examination of the causal role of the tension-resolution structure.

It's also necessary to compare self-selected favorite music with standardized compositions. Perhaps individual preference enhances the effect, or perhaps overly familiar music makes the predictive structure less informative for the brain. The answer is currently unknown.

A third important issue concerns clinical populations. In people with chronic anxiety or post-traumatic stress disorder, stress-regulatory networks may function differently, so the effect observed in healthy young volunteers cannot be automatically generalized to patients.

Finally, long-term studies are needed. It's unknown whether regular music use after stressful events leads to sustained improvements in stress regulation or whether the body adapts over time and the effect diminishes. Long-term studies will determine whether music is merely a short-term recovery tool or whether it can influence more sustainable stress resilience.

The main conclusion

A study by You and colleagues shows that relaxing music after an acute stressful event not only improves subjective mood. Compared to calming natural sounds, it was associated with a faster recovery of several autonomic and endocrine parameters, including heart rate, skin conductance, and cortisol. [28]

However, music did not prevent the stress response itself when played before exercise. Its benefit emerged primarily in the post-stressor phase. This makes the distinction between "protection from stress" and "accelerated recovery" one of the most important findings of the study. [29]

Functional magnetic resonance imaging revealed that recovery was accompanied by a reorganization of interactions between the auditory cortex, insula, and thalamus. The dynamics of these connections tracked changes in musical tension, and individual network changes were associated with improved mood and physiological measures. [30]

Therefore, the most interesting conclusion of the study is not simply that "music is calming." It suggests a specific mechanism: the brain can use the temporal structure of music—the sequential construction and resolution of expectations—as a dynamic signal to rewire stress-regulating networks and return the body to homeostasis. This is a significantly more meaningful neurobiological model of habitual human experience.

News source

You S., Zhang L., Wu G., Du Y. The structural dynamics of music drive acute stress recovery through functional reorganization of stress-regulation networks. Proceedings of the National Academy of Sciences of the United States of America. 2026;123(37):e2611584123. DOI: 10.1073/pnas.2611584123.

This article concerns an original experimental study in humans, not a review or observational analysis of medical databases. A total of 120 healthy volunteers participated in two experiments; the authors combined controlled induction of acute stress with psychological, autonomic, endocrine, and neuroimaging measurements.

The article is listed in PNAS, Volume 123, Issue 37, e2611584123, with an issue date of September 15, 2026; the results were previously published online and publicized on September 8-9, 2026.

The study was conducted by Yi Du's group at the Institute of Psychology, Chinese Academy of Sciences. The first author is Siqi You, co-authors are Lei Zhang and Guowei Wu, and the corresponding author is Yi Du. The work was supported by the Chinese National Science and Technology Innovation 2030 Program - Brain Science and Brain-Inspired Intelligence Technology, Project 2021ZD0201500.