A small fish and a human, hundreds of millions of years apart, build the sensing brain by the same underlying logic. The finding suggests there may be rules a vertebrate brain follows.

Line up the brains of a fish, bird and a mammal, and something unexpected comes up. You do not see three different answers to the problem of making sense of the world. You see one answer, tilted three different ways.

"You can really see it's almost like a continuum," says Emre Yaksi, a professor at the Kavli Institute for Systems Neuroscience in Trondheim. Read across decades of anatomy, the same two ancient pathways carry the world into the forebrain of all these animals. What changes from one to the next is mainly which route does more of the work. Evolution built these brains from different parts, in creatures that parted ways hundreds of millions of years ago. It kept arriving at the same answer anyway.

That is the puzzle the Yaksi lab set out to chase. If animals this far apart on the tree of life keep landing on the same arrangement, perhaps the arrangement is no accident. Perhaps there are organisational rules deep enough that a fish and a person, for all the differences between them, are bond by the same ones. The Yaksi lab’s new study, published in Science with Dr Anh-Tuan Trinh as the first author, is an attempt to catch one of those rules at work in the least likely animal there is.

The problem every brain must solve

Let’s start with what the brain is up against, because everything else follows from it.

Our perception of the world does not arrive whole: it comes in through separate senses. Light through our eyes, sounds and vibrations through others. Each pouring into our brains via on its own pathway. And yet you never experience a jumble of channels. You experience one seamless world. Somewhere inside, the brain has to take those separate streams and merge them back into a single scene. How does the brain arrange itself to manage that?

It helps to picture a house. The senses arrive at the door, and someone has to meet them and show each one where to go. In a mammal, that receptionist job falls to a structure called the thalamus. It receives the incoming senses and sends each to its own room, vision to one, sound to another, keeping them apart at first. Only deeper in the house, in the rooms we call the cortex, do the senses meet again, mingle and get compared, until somewhere in the innermost rooms they become thoughts, perceptions, decisions.

That layout, senses sorted at the entrance and combined in different ways deeper in, is one of the most dependable designs in vertebrate brain evolution. What Trinh, Yaksi and their co-authors wanted to know was whether a creature on a completely different branch of the family tree builds a similar house.

The room

To watch a brain do this, you first have to keep a fish still and content.

A young zebrafish, not yet three weeks old and less than a centimetre long, is settled into a bed of clear gel beneath a microscope. A small opening is made in the gel near its mouth, so fresh water can flow past and it can breathe easily. The fish is then acclimatised to this small, enclosed world, the way a person might be settled and reassured before an MRI scan.

Then there is the wall of instruments. "You activate a whole bunch of switches," he says. "It reminds me of the cockpit of an aeroplane, or a spaceship. There are so many buttons everywhere." He has spent years learning them. "It's like playing the piano. At first, it's very hard. Over time you get better."

What the buttons buy is something he has never stopped marvelling at. He first saw the activity of a living brain more than a decade ago, as a student. "The first time I saw neurons lighting up here and there, it was just like fireworks in the brain. It was so amazing." The fish offers something no mammal can. You do not see a small patch of brain, you see the whole of the forebrain at once, end to end, every neuron flaring the instant it fires, in an animal that is alive and sensing. The entire stage lit and behaving, in real time.


Sorting the world

The experiment itself was simple. Trinh showed the fish a flash of red light. He sent a faint buzz through the water. Sometimes one, sometimes the other, sometimes both together, and watched where the brain answered.

Each signal means something to a fish. A flash of light can be as ordinary as a shadow sliding past, a change in the surroundings worth noticing. The buzz is more pointed. Fish can sense movements in the water, and it is sensitive enough to feel the smallest vibrations. "If a predator comes towards a fish, there's a lot of water movement," Trinh says. The lab's gentle tremor is subtler than that, less an attack than an ambush. "It's really like a surprise signal. Like if somebody sneaks up and taps you on the back. That's the kind of signal we gave the fish."

When the team traced where these signals land, they found the fish keeps a different doorkeeper than we do. It is not the thalamus that meets the senses at the entrance, but another structure altogether, fed from the sensing centres of the midbrain. The researchers call it the PG, which is short for preglomerular complex. PG does the same tidy work. It takes the world in and passes it onward sorted, light towards one region of the forebrain, vibration towards another, each stream still clean and separate. The same first rooms, in a different house.

The cells that wait

But the fish's forebrain does not simply hand the senses along. It works on them, and the deeper Trinh looked, even stranger the cells became.

The plain, single-sense neurons gave way to cells that answered to both light and vibration, the two streams starting to merge. And then, further in, he found a type of neuron he had not been looking for. It stayed quiet when the light flashed on its own. It stayed quiet when the water trembled on its own. It woke only when the two came together at the same moment, and when it did, it fired harder than either event alone could account for.

Anyone who has stood outside in a storm knows the phenomena these cells are built around. The lightning reaches your eyes a moment before the thunder reaches your ears, and still, you know they both belong to the same event. You feel one storm. Something in your head binds the flash to the crack despite that delay. Here in a fish, the researchers caught cells doing exactly that, registering not the flash, not the crack, but the coincidence that the two arrived together.

Trinh found them not at the microscope but afterwards on his computer, deep in the analysis, and he did not believe them at first. "I was blown away. My first reaction was, is this real or not?" He spent the next few hours running the analysis every way he could think of, trying to make the pattern break. It would not. When he finally accepted it, he sat and did nothing useful for a while. "I literally spent ten minutes just looking at these beautiful plots."

What he was looking at was a kind of a hierarchy, ‘a ladder’ built across the brain. Towards the back sat the simple cells, each minding a single sense. Towards the front, the cells that combined and compared. Simple answers near the entrance. Stranger ones more difficult to predict the deeper you went. The same climb from sensing to perceiving that runs through our own cortex. What these front cells are finally for, no one yet knows. The team has watched what they do, not what they are used for, and nobody has yet tested how they shape the way the fish behaves. For now, they are neurons that seem to be built to notice when two things belong together, which is the very thing a brain has to manage before it can learn that one thing causes another.

Why a fish should bother

But why should a tiny zebrafish brain resemble ours at all?

Most of what keeps a mammal alive does not happen in the cortex. The cortex is not for chewing, moving or making babies, and it is not even needed to dodge an obstacle in its path. The forebrain is for the moments when the world stops behaving as expected: a moment where a creature has to find a new way through and when nothing pre-built could have prepared for it. And it seems that the pressure to build such a brain structure, a machine for adapting, pushes very different animals towards similar solutions.

This is where Yaksi reaches, of all things, for soup. "You want to thicken your soup. You can put potatoes and rely on the starch, or you can put flour, and it also works. The solutions are similar, but you do not necessarily need to rely on the same material as long as it functions similarly. Two cooks, different ingredients, one result. Two vertebrate lineages, two different sets of brain parts, one design. Sort the senses, merge them room by room, add in cells that respond to a coincidence, and build upward from there.

Whether the fish inherited this arrangement from a shared ancestor, or arrived at it entirely on its own, is a question the lab is now chasing at the molecular level, comparing the cell types that build the circuit in the fish against those in the mammalian cortex and thalamus. It may turn out the two were assembled from the same raw materials after all. Or the fish may have found its own materials and still ended in the same place. Either way, it is the kind of finding that makes a scientist choose words with care. The fish's doorkeeper is not our thalamus in disguise, and whether the two share any ancient kinship is a question for work still to come. What the fish shows, plainly, is that the road matters less than where it leads.

The rule in the fish
There is an idea behind the work in Trondheim, that the brain runs on discoverable recipes, organising principles it follows the way a kitchen follows a method, and that if you look closely enough you can read them straight off the living tissue. This study catches one of those recipes in the unlikeliest place of all. Not in a primate, not even a mouse, but in the forebrain of a small, transparent fish.

"I don't argue that a fish has the equivalent of a mammalian cortex," Yaksi says. "But a fish has something. It's the pallium. And it evolved from the same vertebrate ancestors that our human cortices evolved from." Most of what it does is still in the dark for him, and he says so gladly. The study, he insists, only found the way in. "We just now learned where the world comes in. That is how everything starts."

But the way in, opens onto something large. If a fish and a person, hundreds of millions of years apart, both take the world in through separate pathways and then stitch it back together by the same logic, then that logic starts to look less like a ‘mammalian accident’ and more like a common rule. Something a brain arrives at again and again, because the task of making sense of a world leaves it little other choice.

A mind, it turns out, can be reached by more than one road. The roads keep ending in the same place.

Reference
Anh-Tuan Trinh, Anna Maria Ostenrath, Ignacio del Castillo-Berges, Fanchon Cachin, Mina Koç, Susanne Kraus, Bram Serneels, Koichi Kawakami and Emre Yaksi, “Hierarchical sensory processing in zebrafish thalamocortical-like circuits,” Science, 2 July 2026. [lim inn DOI]

Funding
The research was supported by the Research Council of Norway, including its Centres of Excellence scheme, a Marie Skłodowska-Curie postdoctoral fellowship (European Commission), and JSPS KAKENHI (Japan).

Ethics
All procedures on zebrafish were carried out in accordance with EU Directive 2010/63/EU and approved by the Norwegian Food Safety Authority.