One of the most earth-shaking results in science occurred with something called the double-slit experiment. In 1803, English scientist Thomas Young used this experiment to "prove" that light wasn't a particle, but a wave. When he sent a beam of light through two parallel slits in a sheet of metal onto a screen, it created an interference pattern — a series of bright bands — that could only exist if the light existed as two interfering waves. If they were particles, after all, the light would just appear as two bands, not several.

The weird part happened around the turn of the century, when scientists were starting to understand that light wasn't just a particle or a wave — it was both. A "photon" is a little packet of light that behaves like a particle, and scientists figured out how to send photons through a double slit one by one. That, of course, should create two simple bands, not an interference pattern — except it didn't. Somehow, the photons still interfered with each other to create that pattern, as if all of the possible paths the photons could take were interfering with one another, even though each photon only took one path.

Weirder still, when scientists placed detectors at each slit to keep track of which one each photon passed through on its way to the screen, the interference pattern vanished. It's as if each photon follows all possible paths until it's observed, at which point it "collapses" into a single path. Weird, right? Well, catch your breath. Things are about to get even weirder.

In the mid-20th century, a physicist named John Archibald Wheeler came up with an interesting variation on the double-slit experiment. When you send a photon through the double slit, he figured, it has one of three possibilities: it can go through the right slit, it can go through the left slit, or it can go through both slits (in which case, it would be behaving as a wave). The double-slit experiment shows that the photons go through both slits when there are no detectors and one slit or the other when detectors are in place.

But what if you made the decision to detect the photons after they already made the "choice" of what path to take? Say you shot the photons at a screen that could be removed at a moment's notice to reveal two detectors, one that would detect a photon that passed through the right slit, another to detect one that passed through the left. A research team tried Wheeler's experiment for real in 2007 (and another extended it into space in 2017), and got the exact results he predicted: If you decide to leave the screen down after the photon passes through the double-slit, you get an interference pattern showing that it took multiple paths. If you decide to remove the screen after the photon passes, it hits either the right or the left detector, showing that it took one path. It's as if the decision to leave or remove the screen actually reaches back in time to change the particle's behavior. As Wheeler put it, "Thus one decides the photon shall have come by one route or by both routes after it has already done its travel."

Of course, many prudent physicists take issue with this interpretation. In fact, the entire premise of the experiment isn't quite right: It's not that a single photon only has one of three possibilities; it's that a single photon embodies all possibilities until it's detected. You see all of those possibilities when the screen is down because nothing was detected. When the screen is removed, the detectors make the photon "collapse" into a state that's detected by one detector or the other — not, like the eerier interpretation suggests, go back in time to "choose" a path.

Even so, the very idea that we might "trick" particles into doing what we want (and having our hopes hilariously dashed as a result) is mind-bending to think about. Quantum physics continues to pose puzzling possibilities about the universe, and that puzzle is what's so cool about it.

For a deeper dive into the quantum world, check out "Quantum Mechanics: The Theoretical Minimum" by New York Times bestselling author Leonard Susskind and Art Friedman. We handpick reading recommendations we think you may like. If you choose to make a purchase through that link, Curiosity will get a share of the sale.