What is Einstein's theory of local realism

"You shouldn't jump to conclusions"

Quantum physics paints an almost unimaginable picture of physical reality. But as successful as it is, not all "loopholes" have been closed - and alternative theories are at least conceivable. One unanswered question is, for example, to what extent the measurements carried out on quanta depend on other physical events. A research collaboration with the help of more than 100,000 volunteers was looking for an answer: They used the human mind as a random number generator to enable measurements that are independent of purely physical objects. In an interview, Wenjamin Rosenfeld from LMU Munich reports how he and his colleagues have closed a loophole with this approach.

World of Physics: You worked on the so-called “BIG Bell Test” together with researchers from a total of twelve institutes. What is the goal behind this?

Wenjamin Rosenfeld

Wenjamin Rosenfeld: Bell tests are about a basic assumption about physical reality. They are named after John Stewart Bell. In 1964, the theorist set up equations with which an old philosophical dispute between Albert Einstein and Niels Bohr about the importance of quantum physics for our view of the world could be decided experimentally in one fell swoop. In our everyday intuition - and also in Einstein's scientific worldview as well as in all of classical physics - “local realism” is predominant.

Can you briefly explain what is meant by local realism?

Simply put, realism means that the properties of physical objects - such as location or speed - are also available without any action. So they are not first brought into being by a measurement or observation. The moon is there, even if no one is looking. And this realism is local because we assume that every particle contains all information about its state: If we measure certain properties on it, then these properties should not depend on distant particles - which cannot causally influence the observed particle . In quantum physics, however, this local realism no longer seems to apply, as Niels Bohr repeatedly emphasized in the dispute with Einstein.

What is a Bell Test?

A Bell test examines the mutual correlations of two or more quantum physically entangled particles. In this state, the particles are very closely linked, no matter how far apart they are: When you measure one particle, the other instantly adopts the same state - they rotate in roughly the same direction when we measure their rotation. This phenomenon can be observed even when no information flow between the two particles is possible. Quantum physics even says that the rotation of the two particles does not have to be unambiguous at all before the measurement. Both particles could have been in an indefinite entangled state. Such correlations are checked in Bell tests.

And what do you examine in the “Big Bell Test”?

Single atom trap

The "Big Bell Test" was about carrying out many different experiments all over the world. For example, one compares the states of rotation - physicists speak of "spin" - or the energy of pairs of light particles. Other research groups have investigated the vibrations of superconducting circuits or the states of atomic clouds. In order to be able to carry out as many different experiments as possible, we now need a large number of random numbers. Because the decision about which property is measured on one particle must not be predictable for the other particle. The random numbers guarantee that we can make many independent measurements that don't have a predictable pattern.

So where does the human mind come into play?

If one assumes that humans have something like free will, then one can use this human freedom of choice to generate more or less unpredictable sequences of numbers. We and the other partners in our collaboration then carry out the experiments with them. That the human mind is free is of course only an assumption and is seen differently by some psychologists or philosophers. However, it may still provide a different kind of unpredictability than the artificial sources used up to now.

Why is unpredictability so important?

Think of these Bell Tests, with their potential loopholes, a bit like a court hearing: Don't jump to conclusions. Is it conceivable that we overlooked something in confirming quantum physics? On the one hand, this can be an imperfection of the measuring apparatus. On the other hand, the sources of the random decisions - typically special random number generators - may appear unpredictable to us, but in reality are related to the properties of the particles. Unfortunately, that cannot simply be ruled out. Then a locally realistic description of the world would not be refuted. Because the quantum physical correlations between the entangled particles would then not even exist. Instead, they would be just an artifact of other, as yet unknown laws. You would have condemned an innocent, so to speak.

How many volunteers took part in the project?

Random numbers by computer game

We're really surprised at how many people have cared about this fairly basic problem. Thanks in part to the good organizational work of the Institute of Photonic Sciences in Barcelona, ​​which led the project, over 100,000 people worldwide took part on November 30, 2016. There was also a small computer game that you could play on your smartphone. The aim was to type in sequences of numbers that were as unpredictable as possible and to compete with others in the process. These sequences of numbers then landed on a server and resulted in a total of 90 million bits of data. Thanks to these human-generated random numbers, all laboratories were then able to carry out independent Bell measurements. As expected, the hypothesis of a locally realistic description of the world was impressively refuted.

Has the last word been spoken and there is no alternative to established quantum theory?

I would put it a little more cautiously: It is still extremely well confirmed and we do not see any plausible alternative with a comparable predictive power. From a strictly logical point of view, however, not all loopholes have been closed. There are a few other loopholes. These could be closed individually with a high degree of certainty, but not all of them at the same time in one experiment. For example, we would like the greatest possible spatial separation between the human random number generators and the experiments. So that any influence is excluded and the test subjects have one second per experiment to make their decision, they would have to be on the moon, for example. But that might take the game a little too much.