Quantum Thought Experiment Works In Space

In Quantum Mechanics by Brian Koberlein

Quantum theory is strange. With it’s particle-wave duality, and spooky action at a distance, it is a difficult theory to wrap your head around. But long before we were able to test some of its strangest implications, we used hypothetical thought experiments. One of the most famous was the delayed-choice experiment proposed by John Wheeler in 1978. 

The delayed choice experiment looks at one of the strangest behaviors of quantum systems, specifically that they can sometimes interact like particles, and sometimes like waves. For example, light can behave like a wave, and waves of light can overlap to create interference patterns that can be seen as ripples of bright and dark on a screen. Light can also behave as particles called photons, and a single photon can trigger a detector, similar to the way a baseball can be used to drop someone in a tank of water.

It would seem, then, that a quantum system such as light is sometimes a wave, and sometimes made of particles. But if that’s the case, you should be able to trick the system. What if you set up a wave experiment for light, and then changed it to a particle experiment after the light has already been released? Or what if you randomly chose a particle or wave experiment long after light began its journey. If the light really has to be either a particle or a wave, then it somehow has to choose before the experiment is decided. What would happen if it chooses wrongly? This is what Wheeler called the delayed-choice experiment, because the choice of experiment is only decided after the light is already committed.

Wheeler’s idea was to imagine a “cosmic interferometer.” Suppose light from a distant distant quasar were to be gravitationally lensed by closer galaxy. As a result, light from a single quasar would appear as coming from two slightly different locations. Wheeler then noted that this light could be observed in two different ways. The first would be to have a detector aimed at each lensed image, thus making a particle measurement. The second would be to combine light from these two images in an interferometer, thus making a wave measurement. According to quantum theory, the results of these two types of experiments (particle or wave) would be exactly as we’ve observed in their standard form. But the light began its journey billions of years ago, long before we decided on which experiment to perform. Through this “delayed choice” it would seem as if the quasar light “knew” whether it would be seen as a particle or wave billions of years before the experiment was devised.

It took nearly 30 years before the experiment was successfully done using an interferometer in a lab. The results were exactly what Wheeler predicted. Even though the choice is delayed, the outcome is just what quantum theory predicts. This was considered solid proof for most scientists, but a few argued that because the experiment was done entirely in the lab, some subtle interaction might have given light a clue as to the outcome. So there has been some effort to do the experiment at much greater distances. Now a team has succeeded in doing it between a lab on Earth and a satellite in space.

Their setup is similar to other delayed choice experiments, but with a spacey twist. Light passes through an interferometer in the lab, but instead of simply measuring the result, the combined beams travel out into space, where they strike a mirror on a satellite in low Earth orbit. Only after the light bounces back to the lab is the outcome measured. The choice of measurement isn’t made until the light beam is well beyond the lab. So the lab can’t affect the outcome. As expected, the result matches Wheeler’s prediction.

So quantum theory is right. Quantum systems don’t choose to be particles or waves to fit the experiment. They have both particle-like and wave-like properties at the same time. It’s strange, but it’s not magic.

Paper: Francesco Vedovato, et al. Extending Wheeler’s delayed-choice experiment to space. Sci. Adv. e1701180 (2017)