When anxiety kicks in, breathing often changes with it. Breaths become faster and shallower, while deliberately slowing them down can have the opposite effect, helping many people feel calmer. That connection is familiar enough that controlled breathing has become a staple of meditation, relaxation exercises, and anxiety management. But exactly how breathing communicates with the brain has been surprisingly difficult to pin down.
A 2026 study published in the Proceedings of the National Academy of Sciences may have uncovered an important piece of that puzzle. Researchers identified what they describe as a “nose-to-brain circuit,” connecting sensory neurons in the nasal cavity with brain regions involved in anxiety. More surprisingly, the circuit appears sensitive to how quickly we breathe through the nose.
That could help explain why breathing and anxiety can sometimes seem to feed into one another. Slow and rapid nasal breathing produced different effects within the newly identified pathway, suggesting breathing may do more than simply respond to our emotional state. The way we breathe could also send signals back to the brain that influence it.
With anxiety disorders affecting hundreds of millions of people worldwide, understanding that connection could have practical implications. The findings don’t mean breathing exercises can replace established anxiety treatments, but they offer a biological explanation for why something as simple as changing the pace of a breath might influence how the brain responds to stress.
The anxiety breathing circuit: what the study found
The research was conducted at Shanghai Pudong Hospital, Fudan University Pudong Medical Center, within the State Key Laboratory of Brain Function and Disorders at Fudan University in Shanghai. Xinsong Guo, Mengyan Liu, and Qingcheng Xiong served as co-first authors, with Ruiqi Wu as corresponding author. Working with male mice, the team used an unusually comprehensive suite of techniques to trace the pathway from the nose inward: “We leveraged optogenetics, chemogenetics, electrophysiology, fiber photometry, behavioral tests, and viral tracing to uncover the cellular and circuit mechanisms underlying this nose-emotion modulation,” the researchers wrote.
The pathway they identified begins with specialized cells inside the nasal cavity. Olfactory sensory neurons (OSNs) detect odorants and send resulting signals to the brain, but research shows they are also mechanosensitive to airflow inside the nose – meaning these neurons physically respond to the movement of air, not just its chemical content. That mechanosensitivity is the foundation of the entire anxiety breathing circuit.
From the OSNs, signals travel to mitral cells in the olfactory bulb, then to parvalbumin-positive long-projecting interneurons in the perirhinal cortex, and subsequently to glutamatergic neurons in the posterior basolateral amygdala, through which nasal afferent activity bidirectionally regulates anxiety in a frequency-dependent manner. That four-stage chain – nasal cavity to olfactory bulb to perirhinal cortex to amygdala – is the anxiety breathing circuit. Each link was not merely observed but causally confirmed through experimental manipulation.
OSN activity triggers corresponding activity in the olfactory bulb, a forebrain structure involved with smell that also connects with the limbic system, a network of brain structures that help regulate emotions and stress responses. The olfactory bulb’s anatomical position at the crossroads of the smell-processing system and the limbic system makes it an ideal relay – and it is precisely why the nose, unlike the mouth, has direct biological access to the brain’s emotional architecture.
The frequency effect: slow vs. fast
The most clinically significant finding in the study is the bidirectional, frequency-dependent nature of the circuit’s effects. Low-frequency nasal airflow or optogenetic OSN stimulation induced anxiolysis (anxiety relief) and increased perirhinal cortex high-gamma power by activating parvalbumin-positive neurons, whereas high-frequency stimulation had opposite effects. The same circuit that slow breathing uses to calm the brain is also the circuit that rapid breathing uses to alarm it.
Nasal airflow regulates anxiety-like behavior through this pathway in a frequency-dependent manner. The effect is also bidirectional, which suggests faster breathing could amplify anxiety just as slower breathing appears to relieve it.
To confirm that the olfactory bulb-to-perirhinal-cortex segment of the pathway was the critical link, researchers chemically silenced it. Chemogenetic silencing of the olfactory-bulb-to-perirhinal-cortex pathway eliminated the frequency-dependent regulation of anxiety-like behaviors driven by OSN stimulation. When that link was severed, the frequency of nasal breathing no longer affected anxiety levels – which is causal evidence that the circuit actively drives anxiety modulation rather than merely correlating with it.
The researchers also tested whether sustained low-frequency nasal airflow exposure could produce lasting effects. A daily brief regimen of low-frequency nasal airflow or optogenetic OSN stimulation persistently reduced anxiety-like behaviors in the anxiety mouse model. Mice placed on a two-week protocol of controlled slow nasal airflow showed both restored neural activity and reduced anxiety-like behaviors over time, suggesting the circuit is not just acutely responsive but potentially trainable.
Why the amygdala is the right target
The posterior basolateral amygdala, where the anxiety breathing circuit ultimately terminates, is not an arbitrary endpoint. The basolateral amygdala has been implicated in threat detection and the generation of anxiety states. Research published in Neuron in 2026 by Ciaran Murphy-Royal and colleagues at the Centre de recherche du Centre hospitalier de l’Université de Montréal confirmed that basolateral amygdala astrocytes reliably encode anxiety states, with astrocyte calcium signaling playing a causal role in driving anxiety behavior mediated by noradrenergic signaling.
The olfactory pathway’s direct access to this structure is neuroanatomically unusual. In humans, as in other mammals, the olfactory system is the only sensory system with direct neural connections to the limbic system without a prior relay in the thalamus. Every other sensory modality – sight, hearing, touch, taste – routes signals through the thalamus before they reach the emotional processing centers. The nose bypasses that relay entirely, sending signals straight to the amygdala via a short, fast path. The 2026 Fudan University study now specifies precisely which neurons and cortical way-stations that fast path travels through when responding to breathing frequency, rather than scent.
This pathway is functionally distinct from the established prefrontal-amygdala circuits in respiration-related fear regulation. That separation means the nose-to-brain anxiety circuit operates in parallel with, not instead of, the conscious top-down pathways that deliberate breathing exercises engage. The two systems may be complementary, and both appear capable of influencing the amygdala’s threat response.
What ancient practice got right
Long before neuroscience could trace individual neurons, cultures across Asia developed breathing practices whose emotional effects were well documented. Humans have long appreciated the emotional effects of breathing, as manifested in deep-rooted cultural traditions such as pranayama, Taoist breathing practices, and various forms of meditation.
Research published in Frontiers in Human Neuroscience in 2025 found that the anxiety-buffering effect of slow breathing may be mediated by enhanced neural activity in the midfrontal region, reflecting improved prefrontal regulatory control over emotional processing. That finding aligns with the new circuit data: if slow nasal breathing activates a pathway that quiets the posterior basolateral amygdala, it would reduce the amygdala’s interference with prefrontal regulatory functions – consistent with the calm, clear-headed state that pranayama practitioners have described for centuries.
A 2020 clinical study in Frontiers in Human Neuroscience found that four weeks of pranayama practice significantly decreased anxiety and negative affect in participants, and that pranayama modulated activity in the amygdala and prefrontal cortex. Those brain regions are precisely the structures that the nose-to-brain circuit identified in 2026 connects and regulates.
The nasal specificity question
One of the most practically relevant questions the study raises is why nasal breathing appears more effective at reducing anxiety than mouth breathing performed at the same rate. Slow nasal breathing appears to reduce anxiety more effectively than slow mouth breathing, pointing to something specific about the nose in emotional modulation.
The 2026 circuit mapping provides a direct answer: mouth breathing does not engage the OSN-to-olfactory-bulb-to-perirhinal-cortex-to-amygdala pathway. Air entering through the mouth bypasses the nasal cavity entirely, so the olfactory sensory neurons never receive the mechanical signal that activates the anxiety-reducing circuit. Nasal breathing rhythm has been shown to synchronize brain oscillations in both humans and rodents, a property that oral breathing does not share.
The olfactory bulb connects with the limbic system, a network crucial for emotional reactivity and stress responses, and the neural oscillations of several brain regions synchronize with nasal breathing cycles. Disrupting nasal inputs through olfactory epithelium or olfactory bulb ablation, or by obstructing nasal airflow, diminishes these respiratory-locked oscillations and elevates anxiety-like behaviors in animal studies. Blocking nasal airflow reduces the calming oscillations and raises anxiety; restoring nasal airflow re-engages the system.
Research involving human intracranial recordings published in Nature Communications in 2026 found that forebrain regions including the amygdala robustly synchronized with breathing rhythms during wakefulness – extending the mouse findings toward a human parallel and reinforcing the idea that the nose-to-brain anxiety circuit is not a rodent-specific phenomenon.
Limitations and the road to human therapy
The study reveals a nose-brain axis that bidirectionally modulates anxiety via nasal afferent frequency, providing potential interventional strategies and targets for anxiety disorders. The authors are careful about how far those implications extend at this stage. The study used only male mice, relied on highly controlled nasal airflow delivered through implanted cannulas, and employed experimental stress procedures that don’t perfectly replicate how anxiety develops in humans. Whether the same four-stage circuit exists in the human nasal cavity, olfactory bulb, perirhinal cortex, and amygdala at the same resolution and with the same frequency sensitivity remains to be established through human trials.
The choice of male-only subjects is also a meaningful gap. Anxiety disorders are roughly twice as prevalent in women as in men globally, so the absence of female subjects limits how far the results can be generalized – particularly for the sex most affected by the disorder being studied. Future research will need to determine whether the same circuit operates identically in female animals before human applications can be designed for all populations.
The mechanistic specificity of the findings is nonetheless unusual for early-stage animal research. Most proposed anxiety circuits rely on indirect correlational evidence. Here, the team chemically silenced a single link in the chain and watched the entire frequency-dependent anxiety effect disappear. That kind of clean causal evidence, replicated through multiple methods, suggests the circuit is real, robust, and worth investigating in human subjects as a priority.
Key takeaways
Scientists have long known that breathing and emotional state are closely connected. What this study adds is a surprisingly detailed look at one way that connection may work. In mice, researchers traced a pathway from sensory neurons in the nose through the olfactory system and ultimately to the amygdala, where changing the frequency of nasal airflow altered anxiety-like behavior.
The discovery offers a compelling biological explanation for why slow nasal breathing may feel calming, but there is still an important gap between mapping this circuit in mice and using it to treat anxiety in people. The researchers used only male mice and highly controlled experimental techniques, and the same pathway still needs to be confirmed in humans.
Still, the findings add to a growing picture of breathing as more than something that simply changes when we’re anxious. The signals created by breathing itself may help shape what happens in the brain. If future human research confirms the same pathway, something as ordinary as breathing through the nose could give scientists a new window into how the brain regulates anxiety.
Disclaimer: The information provided here is for educational and informational purposes only and is not a substitute for professional psychological, psychiatric, or mental health advice, diagnosis, or treatment. Always seek the guidance of a licensed mental health professional, therapist, psychologist, or psychiatrist with any questions or concerns about your emotional well-being or mental health conditions. Never ignore professional advice or delay seeking support because of something you have read here.
AI Disclaimer: This article was created with the assistance of AI tools and reviewed by a human editor.
Read More: Researchers Discover Possible Way to ‘Turn-Off’ Anxiety Signals to the Brain