Loading today’s stories
Science1 distinct publisher3 min readPublished
Researchers at DIfE and Charite paced breathing at a two-to-eight inhale-exhale ratio and saw bolder bets alongside stronger reward-region activity. The release names the direction of the effect but not its size.
The Scientist · Science desk
Compiled by The ScientistSomething wrong?How this is made
The useful feature of this design is that the breathing was assigned rather than observed [2]. Watch people breathe while they gamble and you learn little about direction, since someone who is agitated breathes fast and also plays it safe. Pacing the breath with visual cues, and using each participant's own resting rhythm as the comparison condition, puts the manipulation upstream of the choice [2]. That earns a causal claim about breathing and risk taking. It does not earn one about heart rate variability, which the release reports as increased under the long exhale [7] but which was recorded while the breathing changed rather than set independently, nor about the ventromedial prefrontal cortex signal [6]. The work was led by Soyoung Q Park at the German Institute of Human Nutrition Potsdam-Rehbruecke, with collaborators at Charite, Freie Universitaet Berlin and the German Naval Institute of Maritime Medicine [8].
The asymmetry is the part worth sitting with. According to the release, the shift came from greater weight on possible rewards, with losses treated much as before [5]. The mechanism is specific rather than a general calming story, and it is also the fragile half of the result: a preserved loss response is an absence of a detected difference, and the summary published by the German Center for Diabetes Research gives no effect size for either component [11][13]. In 41 people [1], the defensible reading is that reward weighting moved enough to detect and loss weighting did not, which is not the same as loss weighting holding still.
The protocol as published is a ratio, not a schedule. Eight over two means the exhale ran four times as long as the inhale [12], but no cycle length is given, so it cannot be converted into breaths per minute or copied by anyone who wants to try it [15].
Alongside the fMRI, the team recorded heart activity, skin conductance and pupil responses, which is the right instrument set for asking whether a long exhale does more than lower heart rate [3]. The release does not report what those channels showed [16]. Pupil diameter and skin conductance are the standard checks on a generic-arousal account: if slow breathing simply lowered sympathetic tone across the board, a reward-specific story is harder to hold.
Lead author Wenhao Huang describes the interplay between breathing and cardiac dynamics as making the brain more receptive to rewards [10], and the group places the work within neurovisceral models of body-brain communication [14]. What it does not tell you is how long the bias outlasts the pacing, or whether it appears when nothing on a screen is cueing the breath. The release also does not say whether the gambles paid real money [17]. Lying in a scanner following a visual metronome is some distance from a decision with a counterparty.
Take it as this: an assigned four-to-one exhale moved 41 people's choices toward the reward side of the ledger, in the direction the researchers predicted, by a magnitude not yet in public view [1][4][12][13].
Ranked by verification strength, evidence, and original report placement.
Researchers studied 41 healthy participants while they made risky choices under controlled breathing conditions; the findings were published in Neuron.
Participants followed visual breathing instructions and either breathed at their normal individual pace or slowed their breathing while extending the exhalation phase, at a 2:8 inhale-exhale ratio.
Functional MRI tracked brain activity while breathing patterns, heart activity, skin conductance and pupil responses were monitored at the same time; the combination was intended to test whether extending the exhale did more than simply lower heart rate.
Extended exhalation slowed heart rate and was associated with riskier choices.
The change appeared to come from greater attention to possible rewards rather than reduced concern about potential losses; participants still responded to losses in a similar way.
The researchers detected stronger activity in the ventromedial prefrontal cortex and the precuneus, regions involved in reward processing and in regulating the timing between heartbeats, a measure known as heart rate variability.
Distinct publishers with included, body-backed reporting in this cluster.
1 article · September 4, 2026
Follow any of these and your For You feed starts watching them — no settings page required.
science
A single CD19 CAR-T infusion held three of six refractory rheumatoid arthritis patients drug-free2 distinct publishers
science
One enzyme sets the chromatin mark that links early stress to adult anxiety in mice1 distinct publisher
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
One institutional announcement, direction only
Every fact here traces to a single press announcement issued by the German Center for Diabetes Research and reproduced by ScienceDaily; the Neuron paper it summarises is not part of our coverage. The text gives a mechanism and a direction but no numbers: no effect size for the shift toward risk, no variance, and nothing from the skin conductance and pupil channels it says were recorded.
Nothing deployed to observe
This is a scanner study of 41 people. The uses the researchers float — everyday self-regulation, a non-pharmacological adjunct in anxiety or depression, breath work applied to eating behaviour — are stated as future research, and no trial, clinic or product follows from them in anything we have.
'Scientific proof' from 41 volunteers
Park's quote claims the study provides scientific proof of a reliable and targeted method capable of controlling our decisions. The result behind that sentence is an association in 41 healthy adults in one session, reported without an effect size, and the announcement's own headline retreats to 'may push'. What is overstated is the certainty language; the finding itself is more modest.
The institute publicising its own study
The announcement comes from the research network the authors' institute belongs to, quotes only the authors, and closes by proposing the next study on overweight populations and food-related choice — squarely the remit of a nutrition institute inside a diabetes research network. Federal grant numbers are printed at the foot, which is candid about who paid, and no independent voice is present to weigh the interpretation.
Clear provenance, unverifiable numbers
Who did the work, where, and in which journal are all unambiguous, and the internal account is coherent, which is why this is not lower. It stays under halfway because the findings that matter — riskier choices, an intact loss response, stronger reward-region activity — can only be taken on the institutes' word until the Neuron paper is read, and basic task detail such as whether real money was at stake is missing.