A decay worth forty years of waiting, and the gut signal that teaches you what to like

Read the episode (Narration script)

Some results arrive with a bang. This one arrived as a small pile of events that almost nobody would notice, except that physicists have been waiting for them since the 1980s.

The LHCb collaboration at CERN has just posted the first evidence, from a single experiment, for a decay called B-zero to two muons. A B meson is a short-lived particle the Large Hadron Collider produces in enormous numbers. Nearly all of them come apart in ordinary ways. But roughly once in ten billion times, a B-zero should turn into nothing but a pair of muons, the heavy cousins of the electron.

And the reason that number is so small is rather lovely. The Standard Model, the theory covering all the known particles, does not allow this decay directly at all. It can only go by a detour, through heavier particles the B briefly borrows and gives straight back. So the rate comes out tiny, and the theory predicts it very sharply. Which makes it one of the best places in physics to catch the theory being wrong. If any unknown particle is lurking in that detour, the rate should not match.

It matched. Using collisions recorded through 2024 and 2025, LHCb measures about 1.1 of these decays per ten billion, against a prediction of about 1.0. They also made the world's most precise measurement of the closely related B-s decay: about 3.25 per billion, where the theory says 3.66. Everything agrees with the Standard Model.

The B-zero result sits at about three standard deviations, which particle physicists call evidence rather than discovery. The bar for discovery is five. Still: forty years of looking, and the thing is finally there, behaving itself. There is something quietly funny about a field where the most spectacular possible outcome would have been to be wrong.

From the very small to the dinner table. When you learn to like a food, two separate judgments are involved. Your mouth gives an instant opinion: sweet, salty, strange. Then, much more slowly, your gut reports back on whether there were any actual calories in it. The brain has to staple those two messages together, even though they arrive many minutes apart.

A team at the University of California, San Francisco, with colleagues at McGill and in Cologne, has published in Nature the pathway that does the stapling. In mice, nutrients arriving in the gut trigger a release of dopamine, but not in the usual reward territory of the striatum. It lands instead in one small patch of the amygdala, toward the front, and preferentially on the left. Silence those neurons and the mouse eats perfectly normally. It simply never learns that the flavour predicts the calories. And when the researchers looked at human brain imaging, that same left-side region was the one that responded once nutrients reached the gut.

So the reason you crave a particular meal is not only taste. It is a slow confirmation signal from your intestines, delivered to a region better known for fear, arriving long after the plate is empty.

Also in Nature this week, a far less glamorous result that may touch more lives. High blood pressure is the most common serious condition people do not know they have; more than half of adults who have it are unaware. Researchers at Google Research, with Beth Israel Deaconess Medical Center and the University of Washington, tested whether an ordinary consumer smartwatch could flag it with no cuff involved at all, reading only the optical pulse sensor at the wrist and the motion sensor.

A hundred and ninety-six people wore the watch for a week and were checked against twenty-four hour ambulatory monitoring, the gold standard. The watch picked up about two thirds of cases, with ninety per cent specificity. A single blood pressure reading taken at a clinic visit picked up about fifty-five per cent. That is not a diagnosis, and nobody is claiming it is. It is an invitation to go and get one.

A quick one from the deep past of plants. Every land plant descends from freshwater green algae, and the closest living relatives are the zygnematophytes, the group responsible for the slimy green threads in a pond. Researchers led from Gottingen sequenced forty-three new strains and built a family tree spanning more than a hundred species. The common ancestor was almost certainly a single cell, and the threads, which is to say multicellularity, the trick that eventually made forests possible, evolved more than once, independently. Not one staircase upward. Several separate attempts, running in parallel.

And finally, a picture. Astronomers using LOFAR, the low-frequency radio array spread across Europe, have released the deepest radio image ever made at these frequencies: two hundred hours of observing, four hundred terabytes of data, a patch of sky about twenty-five times the area of the full moon, and nearly sixty-three thousand radio sources sitting in it. The resolution is about a third of an arcsecond, sharp enough to separate the jets of a supermassive black hole from the galaxy around it.

The real news is buried in the method. The calibration finally got good enough, and cheap enough, that an image like this is now something you can simply decide to make. Which is often how a frontier actually moves. Not a leap, but the morning the impossible becomes merely expensive.