No one really knows what happens inside an atom. But two competing groups of scientists think they've figured it out. And both are racing to prove that their own vision is correct.

Here's what we know for sure: Electrons whiz around "orbitals" in an atom's outer shell. Then there's a whole lot of empty space. And then, right in the center of that space, there's a tiny nucleus — a dense knot of protons and neutrons that give the atom most of its mass. Those protons and neutrons cluster together, bound by what's called the strong force. And the numbers of those protons and neutrons determine whether the atom is iron or oxygen or xenon, and whether it's radioactive or stable.

Still, no one knows how those protons and neutrons (together known as nucleons) behave inside an atom. Outside an atom, protons and neutrons have definite sizes and shapes. Each of them is made up of three smaller particles called quarks, and the interactions between those quarks are so intense that no external force should be able to deform them, not even the powerful forces between particles in a nucleus. But for decades, researchers have known that the theory is in some way wrong. Experiments have shown that, inside a nucleus, protons and neutrons appear much larger than they should be. Physicists have developed two competing theories that try to explain that weird mismatch, and the proponents of each are quite certain the other is incorrect. Both camps agree, however, that whatever the correct answer is, it must come from a field beyond their own.

Related: The Biggest Unsolved Mysteries in Physics

Since at least the 1940s, physicists have known that nucleons move in tight little orbitals within the nucleus, Gerald Miller, a nuclear physicist at the University of Washington, told Live Science. The nucleons, confined in their movements, have very little energy. They don't bounce around much, restrained by the strong force.

In 1983, physicists at the European Organization for Nuclear Research (CERN) noticed something strange: Beams of electrons bounced off iron in a way that was very different from how they bounced off free protons, Miller said. That was unexpected; if the protons inside hydrogen were the same size as the protons inside iron, the electrons should have bounced off in much the same way.

At first, researchers didn't know what they were looking at.

But over time, scientists came to believe it was a size issue. For some reason, protons and neutrons inside heavy nuclei act as if they are much larger than when they are outside the nuclei. Researchers call this phenomenon the EMC effect, after the European Muon Collaboration — the group that accidentally discovered it. It violates existing theories of nuclear physics.

Or Hen, a nuclear physicist at MIT, has an idea that could potentially explain what's going on.

While quarks, the subatomic particles that make up nucleons, strongly interact within a given proton or neutron, quarks in different protons and neutrons can't interact much with each other, he said. The strong force inside a nucleon is so strong it eclipses the strong force holding nucleons to other nucleons.

"Imagine sitting in your room talking to two of your friends with the windows closed," Hen said.

The trio in the room are three quarks inside a neutron or proton.

"A light breeze is blowing outside," he said.

That light breeze is the force holding the proton or neutron to nearby nucleons that are "outside" the window. Even if a little snuck through the closed window, Hen said, it would barely affect you.

And as long as nucleons stay in their orbitals, that's the case. However, he said, recent experiments have shown that at any given time, about 20% of the nucleons in a nucleus are in fact outside their orbitals. Instead, they're paired off with other nucleons, interacting in "short range correlations." Under those circumstances, the interactions between the nucleons are much higher-energy than usual, he said. That's because the quarks poke through the walls of their individual nucleons and start to directly interact, and those quark-quark interactions are much more powerful than nucleon-nucleon interactions.

These interactions break down the walls separating quarks inside individual protons or neutrons, Hen said. The quarks making up one proton and the quarks making up another proton start to occupy the same space. This causes the protons (or neutrons, as the case may be) to stretch and blur, Hen said. They grow a lot, albeit for very short periods of time. That skews the average size of the entire cohort in the nucleus — producing the EMC effect.

Related: Strange Quarks and Muons, Oh My! Nature’s Tiniest Particles Dissected

Most physicists now accept this interpretation of the EMC effect, Hen said. And Miller, who worked with Hen on some of the key research, agreed.

But not everyone thinks Hen's group has the problem worked out. Ian Cloët, a nuclear physicist at Argonne National Laboratory in Illinois, said he thinks Hen's work draws conclusions that the data doesn't fully support.

"I think the EMC effect is still unresolved," Cloët told Live Science. That's because the basic model of nuclear physics already accounts for a lot of the short-range pairing Hen describes. Yet, "if you use that model to try and look at the EMC effect, you will not describe the EMC effect. There is no successful explanation of the EMC effect using that framework. So in my opinion, there's still a mystery."

Hen and his collaborators are doing experimental work that is "valiant" and "very good science," he said. But it doesn't fully resolve the problem of the atomic nucleus.

"What is clear is that the traditional model of nuclear physics … cannot explain this EMC effect," he said. "We now think that the explanation must be coming from QCD itself."

QCD stands for quantum chromodynamics — the system of rules that govern the behavior of quarks. Shifting from nuclear physics to QCD is a bit like looking at the same picture twice: once on a first-generation flip phone — that's nuclear physics — and then again on a high-resolution TV — that's quantum chromodynamics. The high-res TV offers a lot more detail, but it's a lot more complicated to build.

The problem is that the complete QCD equations describing all the quarks in a nucleus are too difficult to solve, Cloët and Hen both said. Modern supercomputers are about 100 years away from being fast enough for the task, Cloët estimated. And even if supercomputers were fast enough today, the equations haven't advanced to the point where you could plug them into a computer, he said.

Still, he said, it's possible to work with QCD to answer some questions. And right now, he said, those answers offer a different explanation for the EMC effect: Nuclear Mean-Field Theory.

He disagrees that 20% of nucleons in a nucleus are bound up in short-range correlations. The experiments just don't prove that, he said. And there are theoretical problems with the idea.

That suggests we need a different model, he said.

"The picture that I have is, we know that inside a nucleus are these very strong nuclear forces," Cloët said. These are "a bit like electromagnetic fields, except they're strong force fields."

The fields operate at such tiny distances that they're of negligible magnitude outside the nucleus, but they're powerful inside of it.

In Cloët's model, these force fields, which he calls "mean fields" (for the combined strength they carry) actually deform the internal structure of protons, neutrons and pions (a type of strong force-carrying particle).

"Just like if you take an atom and you put it inside a strong magnetic field, you will change the internal structure of that atom," Cloët said.

In other words, mean-field theorists think the sealed-up room Hen described has holes in its walls, and wind is blowing through to knock the quarks around, stretching them out.

Cloët acknowledged that it's possible short-range correlations likely explain some portion of the EMC effect, and Hen said mean fields likely do play a role as well.

"The question is, which dominates," Cloët said.

Miller, who has also worked extensively with Cloët, said that the mean field has the advantage of being more well-grounded in theory. But Cloët hasn't yet done all the necessary calculations, he said.

And right now the weight of experimental evidence suggests that Hen has the better of the argument.

Hen and Cloët both said the results of experiments in the next few years could resolve the question. Hen cited an experiment underway at Jefferson National Accelerator Facility in Virginia that will move nucleons closer together, bit by bit, and allow researchers to watch them change. Cloët said he wants to see a "polarized EMC experiment" that would break up the effect based on the spin (a quantum trait) of the protons involved. It might reveal unseen details of the effect that could aid calculations, he said.

All three researchers emphasized that the debate is friendly.

"It's great, because it means we're still making progress," Miller said. "Eventually, something's going to be in the textbook and the ball game is over. ... The fact that there's two competing ideas means that it's exciting and vibrant. And now finally we have the experimental tools to resolve these issues."

Originally published on Live Science.