The 9 Experiments That Will Change Your View of Light (And Blow Your Mind)

There is much of science that we understand. If I threw a ball into the air, and I was given the right data about the forces acting on it, I could tell you exactly where it would land. Science explains through chemistry the molecules that make the ball up. We can predict the energy levels of the sound it would make when it lands. Much like a candle being held up in the dark, science illuminates our view of the world around us… but there is a limit to how far the light currently falls. Even today

, when it feels like there is so much of the world and the universe that we can explain, there is darkness, too. Answers we don’t yet have, and worse; confusing results that erode our confidence in what we think we do know. There are experiments that seem to suggest that light is lying to us, and call into question the very nature of reality. Are we real? Is time linear? …Perhaps not. But are you ready for the comforting veil of understanding to be torn away, and for the strangeness at the edges

of our understanding to be brought into the light? If so, have I got some experiments for you. I’m Alex McColgan, and you’re watching Astrum. And in today’s supercut, I will show you nine experiments that will challenge your understanding of the fundamental laws of physics in a way that will almost certainly leave you with something between a headache and existential dread. You have been warned. And curiously, almost all of these experiments have something to do with light. Light is so much str

anger than you might think. Sure, it may seem simple enough, travelling around the universe delivering energy from one place to another. It helps us see. It provides life to plants, and thus to our planet generally. It has a reputation for being very fast. And yet, for a source of energy that has become synonymous with greater understanding, light is surprisingly difficult to understand. Light helps us see other things better, sure, but when scientists tried to look at light itself, it was surpr

isingly difficult. No, I don’t mean that they started staring into any lamps – please, don’t do that at home - but experiments in the last 200 hundred years or so have proven that what light appears to be and what light is are actually two different things. The first experiment on our list highlights the following mystifying fact: light behaves differently when you’re not looking at it, compared to when you are. But to understand that, let’s begin with the basics. What is light? In the early 170

0’s Isaac Newton theorised that light was made up of tiny little particles that he called “corpuscules”, but in 1801 – nearly 100 years later - a man named Thomas Young discovered that light must actually be more wave-like than particle-like. He proved this using an important method known as the double slit experiment. He set up a source of light, and shone it through two narrow slits onto a board. Young noticed that rather than getting two bands of light on the other side of the slits, a strang

e striped pattern was forming. This was known as an interference pattern, and was incontrovertible proof that light had been travelling as a wave. Why? Let’s talk about waves for a moment. When waves travel, they oscillate up and down. But when two waves try to oscillate the same point in space at the same time, you get something known as interference. Imagine you had a bathtub with a rubber duck sitting on the surface. Two waves reach the duck at once. One wave tries to raise the duck up at the

exact same time the other wave tries to drop it down. What happens? Provided the waves are of the same magnitude and are perfectly out of phase, they will cancel each other out, and the duck would not move at all. This is called destructive interference. Similarly, if the waves both tried to raise the duck up at the same time, the duck would be raised twice as high. This is known as constructive interference. Because waves tend to expand in a circle, two waves next to each other will start to b

oth constructively and destructively interfere with each other. Here are two waves in water. See these lines? These calmer patches are where the waves are cancelling each other out: This is the effect we see with light travelling through the two slits. As the light from one slit propagates, it cancels out the other wave of light at certain points, creating the interference pattern that Young noticed on the board. So, the mystery was solved. Light was a wave, and not a particle. Except, there is

more to this experiment than meets the eye. Let’s fast forward another 100 years, to 1905. Scientists around this time had become puzzled by something known as the photoelectric effect. It turned out that when you shone a light on a metal surface, electron-like particles were coming off it. This was deduced to be because electrons in the metal were getting knocked off it by the increased energy the light was imparting. Imagine it like fruit on a tree. If you pull the fruit off the tree, you need

to use a certain amount of energy. Once the energy is greater than the strength of the fruit’s connection to the branch, the fruit pops off. This was happening with the light and the electrons. Once the light hit an electron and gave it enough energy to pass the threshold, it broke free from the metal. However, what surprised scientists was that if you increased the intensity of the light, they had expected the electrons to be knocked away faster. If you pulled the fruit off the tree harder, it

would come off faster. More energy = more departing kinetic energy. However, this did not appear to be the case. Instead, increasing the frequency of the light increased the velocity of the departing electrons. The intensity of the light didn’t affect the departing electrons’ velocity at all, but did affect the quantity of electrons being emitted. This was a bit of a puzzler. Albert Einstein was the man who solved the puzzle. He deduced that light must be travelling in little packets of energy,

so sending more of them – increasing the frequency – was the only way to increase the energy going to the electrons. He called these packets photons, and later earned a Nobel prize for his work. Light, it seemed, was more like a particle again. Or both a wave and a particle at once? Of course, even this is not the full picture. To be honest, we aren’t completely sure about the full picture even now. Instead, we have more results that are contradictory. Let’s go back to the double-slit experimen

t. Armed with the knowledge of photons, physicists once again took a look at the double-slit experiment. Experimental techniques had improved in the last 100 years, and it was now possible to emit a single photon of light at a time. So, the double-slit experiment was done again. This time, only a single photon would be sent through the slit, onto a detector on the far side. When this was done, the detector registered the arrival of the photon at just a single point. So, light was behaving like a

particle again. But then, why had it interfered with itself in the previous version of the experiment? Scientists had an idea. They sent through multiple photons, one at a time, and plotted the results on the detector. And this is where the result became really strange. Once again, the detector started seeing the photons arriving at single points, one at a time. But bafflingly, the arriving photons started creating a pattern: It was the interference pattern. The proof that light behaved like a

wave. But strangely enough, this was occurring only when a single photon was going through at a time. Somehow, the single photon – which was leaving the detector like a particle and was arriving at its destination as a particle – was apparently in some way travelling through both slits at once, enough to then interfere with itself on the other side, like a wave. If light was just a particle, then when it went through the slits, you wouldn’t see this pattern. You would see only two blobs of light

– one for particles that went through the one slit, and one for particles that went through the other. And yet, here was the interference pattern with its multiple lines of light, disproving that. Scientists tried to pin light down. They set up the experiment, but this time with two more detectors at the slit, so that scientists could observe whether it was indeed passing through both at the same time. It didn’t. But at the same time, it stopped creating an interference pattern on the furthermo

st detector. And from this, scientists began to realise something. Light cared about being observed. To be clear, it didn’t matter whether it was observed by a human eye or a machine. The moment light was interacted with in some way, by any particle – which is the only way we can detect light, there’s no other way to observe it - it started behaving differently than if it hadn’t been detected at all. It was as if light was snapping into focus any time the universe asked it the question of where

exactly it was, when without that scrutiny it appeared to relax into something a little more nebulous. Bizarrely enough, to me this seems to imply that light actually is more like a wave of probability, rather than any discrete particle or wave. Any time it was asked where it was, it confidently provided a definitive answer – it WAS at this point on the detector, it WAS NOT at any other point. But with no-one checking up on it, light seems to be travelling in all directions at once, in accordanc

e with certain probabilities. If you ran the experiment multiple times, you could quantify those probabilities, discovering that it was more likely to be on the bands of the interference pattern, and less likely to be in the gaps. But any time a single photon of light was asked, it gave an answer that was 100% concrete. This is highlighted through something known as the three-polariser paradox. Consider for a moment a pair of polarising sunglasses. Obviously, these reduce the amount of light tha

t can pass through them; usually by about 50%, depending on the type of lens and the wavelength of light. They work by being formed of thin chains of molecules that run lengthways across the lens. Any light that oscillates in the same orientation as this lens gets absorbed. Any that is perpendicular to the chains can pass through without trouble. The interesting case occurs when a single photon is passed through in an orientation that’s diagonal to the lens. In this case, you don’t get half a ph

oton going through. Apparently you can’t just absorb the part of the oscillation that is parallel to the lines and let through the part that is perpendicular. Instead, the photon “snaps” into either the one orientation, or the other. It either is completely absorbed, or passes through entirely – but now with a new, perpendicular polarisation, to match what it would have had to have been to pass through easily. How do we know that the photon wasn’t this orientation all along? Because of what happ

ens when you start adding more lenses. When you place a second lens behind the first, you can block out the light entirely, provided the two polarisations are perpendicular to each other. Let’s say, we rotate the second lens 90 degrees compared to the first one. Any light that gets through the first lens has a 0% chance of getting through the second, like trying to post a letter through a chain-linked fence. As a result, we only see black. But add a third lens, and place it at a 45-degree angle

between the other two, and bizarrely light starts making it through all 3 lenses again. This may seem counter-intuitive – how does adding more blockages increase the amount of light that makes it through? But this result actually rules out the possibility that the light has a fixed orientation. It must be snapping into focus at each new lens, rolling a quantum dice each time to see if it was the right orientation all along or not. If it makes it through the first lens (a 50% chance), it only did

so because it was oriented perfectly perpendicular to the lens’s polarisation. Which means once it reaches the second, it’s coming at it from a polarisation that’s diagonal. So, once again there is a 50:50 chance that it makes it through. It rolls its quantum dice again, and once again has a 50:50 chance of proceeding. If it gets through this hurdle too, then it again snaps to the new orientation, as if it were that new orientation all along (which it obviously wasn’t). Which means that it’s no

w polarised diagonally relative to the third lens, meaning that it now has a final 50% chance of getting through. Of course, some photons do not make it through all 3 of these probabilistic gauntlets. Only about 12.5% of them make it. But that’s more than 0%, which is what was happening previously when you had only two lenses. Light likes to behave in discrete quantities. It is “quantum”. It seemingly snaps to a discrete value when observed. And honestly, we don’t really know why. If you think a

bout a wave, there is no reason why you couldn’t simply have half a wave. You could halve it again and again an infinite number of times and still have an answer that makes mathematical sense. And yet it seems that down on a low enough quantum scale, you can’t halve light past a certain point. You can’t have half a photon, or even one and a half photons. And if you try to do so, the photon instead snaps to one or the other nearest integer, based on probabilities: but only when it’s asked. Otherw

ise, it’s quite content to exist probabilistically, interfering with itself like a wave as it travels along, before jumping to an answer when later asked exactly where it is. What is going on here? This is still being theorised about. The closest comparison we have to it is something known as harmonics, where on a bounded string, only a certain number of waves can exist. On a guitar string, you can have one wave, or two, or more, but never any number that isn’t a whole number. It seems that ligh

t works in the same way. Perhaps something pinches the beginnings and the end of the path light travels down – although what this might be, and what mechanisms drive it, are unknown as of now. Fundamentally, though, perhaps the craziest thing about all of this is that this isn’t just about light. Although we’ve focused on light behaving like a wave, and behaving probabilistically, all particles of matter do the same. Light is just another form of energy, and energy and matter are linked. Particl

es of matter – atoms and even complex molecules – have been shown to have wavelengths. Electrons are just as quantifiable and just as driven by probabilities as photons are. We are apparently all driven by probability, if you scale things down small enough. So, what is everything truly made of? What makes up energy and matter, that causes it to behave in the way that it does? What Is going on under the hood of reality? Why is the universe behaving different when looked at compared to when not? A

nd what does it imply to think that even you are on some level probabilistic? What this all means is anyone’s guess. The person who figures it out will be the Einstein of our time. But for now, all we can say is that when it comes to reality, it seems the universe is playing dice. You and the world around you might be a lot less certain than you might have thought. So completes our first few experiments highlighting the strangeness of light. Take a breather for a moment. Give your brain a chance

to unknot itself. From here-on in, it’s only getting weirder. If there’s one thing I’ve learned about light, it’s that for unthinking energy, light seems to love messing with us. As I just showed you, scientists debated about whether it was a particle, or a wave, because it keeps exhibiting elements of both, seemingly unable to settle. Bizarrely, it behaves one way when you’re looking at it, but a different way when you’re not. But at least its speed is consistent. Light travels at the speed of

light. No matter your frame of reference, that one thing remains the same... I have some bad news for you. It turns out the constancy of light’s speed might not be right either, and the next few experiments I’m about to show you proves it. Light might go slower than physics would predict in certain circumstances. And no, I’m not just talking about light slowing down in denser mediums like glass, although that’s what I originally intended this video to be about. We have an explanation for that.

I’m saying that in some circumstances, light seems to travel a path through time and space that has it either going slower or faster than the speed of light, even if dense mediums aren’t present. But the really weird thing is that it ends up at the same destination in time and space anyway. Let me show you what I mean. Light travels at 299 792 458 m/s. According to relativity, this is the only speed light can travel at, and interestingly seems to stick to that number regardless of your frame of

reference. Two people could be travelling through space – one at 1% the speed of light, and the other at 50% the speed of light – but if they both look at the same beam of propagating photons, they will see them travelling at the same speed. Time and distance would seemingly rather warp than allow you to see anything other than light travelling at light speed. Of course, when scientists say this, they are only talking about light travelling in a vacuum. We’ve known for a long time that as soon a

s you get matter involved, light gets bogged down and travels slower. Light travelling in air only goes at 299,705,000 m/s, a full 87,458 m/s slower than light in a vacuum. Light in water goes around 225,000,000 m/s. Light going through glass caps out at around 200,000,000. The reasons for this are intriguing, but fairly well understood, and certainly not physics-breaking. When light travels through matter, its constantly waving electromagnetic fields gets the electrons within the matter to also

start moving, like ships bobbing on water. But, as electrons moving up and down also generate an electric field that in turn creates a magnetic field, a second light wave is created by these moving particles that crucially overlaps the waves of the original light, albeit one that waves at a slightly different pace to the original light (exactly what speed varies depending on the material). When two waves meet, they interfere with each other – they take an average, sometimes interfering construc

tively to build each other up, and sometimes working against each other. So, when you take the grand total of all the ups and downs of each wave, you actually end up with a new wave – one that travels at a different speed to the other two, and one that goes slower than the speed of light. Eventually, this propagating wave can reach the edge of the blocking material and, without those electrons interfering anymore, you’re left with just the original light again, which is then free to travel along

its original path again at its original speed as if nothing had ever happened. Scientists have had a lot of fun with this concept over the years.Researcher Lene Hau at Harvard in 1999 was able to slow down light to an astonishing 61 km/h by sending it through a cloud of sodium atoms that had been cooled to 1 billionth of a degree above absolute zero. 2 years later, Hau managed to slow down light’s speed to 0, before warming up the cloud and sending it on its way again. You might find that resul

t surprising. However, strange things have happened in the opposite direction too. In 2000, researchers at the NEC research institute in Princeton, New Jersey sent a pulse of light through a cloud of caesium atoms. Alarmingly, when they timed to see how quickly the pulse exited the cloud, it seemed that it exited before it had entered. While this might appear to mess with causality – how can you leave a building before you go inside it, after all? – fortunately there was a simple explanation tha

t saved us from creating too many paradoxes. Although the light pulse travelled faster than light, the light itself did not. This was more an optical illusion than a refutation of Einstein’s relativity. Let’s take a closer look at a photon of light. As Einstein showed us, each photon represents a tiny packet of waves, moving up and down. The speed the waves inside the packet propagate is known as its phase velocity, while the speed at which the packet as a whole is travelling is known as the gro

up velocity. You can also have a wavefront velocity, which is how fast the first photon in a wave of photons can travel. This is a little heavy in its terminology, so let’s explain it with an example. Think of a crowd of people doing a Mexican wave. The wave that the people are doing is the phase velocity. You can see the wave travelling along through the crowd, it might look like it’s travelling quickly. But the crowd itself isn’t going anywhere, so our wave’s true speed is 0. These people are

the group velocity, or possibly the wavefront velocity. Let’s imagine that we wanted to send our crowd marching. They could do so, and could keep doing a Mexican wave as they travelled. But although their waving hands might make the wave go really fast in the direction of their travel, it would vanish whenever it reached the front of the crowd. Information exchange couldn’t go faster than the walking speed of the crowd itself, regardless of how fast the peaks in the waves seemed to be travelling

. Einstein in relativity never claimed that phase velocity couldn’t exceed light speed. He just claimed that information couldn’t travel faster than light. And if you’re trying to deliver a message to someone by sending a crowd of Mexican wavers in their direction, it really doesn’t matter how fast they’re waving. Until the first person in the crowd arrives, no information can be delivered. Still, this difference between the waves within light and the speed of light itself will become interestin

g in our next experiment. And this is where things start to get a little weird. Oh, you thought it was weird already? Oh, no, this is the really physics-defying part. Let’s think back on the double slit experiment. There, researchers explored how light can sometimes behave like a wave, and sometimes like a particle. However, in 2023 researchers from the Imperial college figured out a way to separate the slits of this experiment, not in space, but in time. The way they did this was simple. They t

ook a transparent material called indium-tin-oxide, that under specific conditions can be made to be reflective. Indium-tin-oxide is the stuff they use in most mobile phone screens. They fired a laser at it, and then rapidly changed the material from transparent to reflective and then back again. This left only a slim window – a few femtoseconds – where the laser was reflected. They called this a “time slit”. They recorded what the laser looked like after it had been reflected, and found that it

s frequency had spread out a little bit in the process, but other than that nothing too crazy had happened. The weird thing was what happened when they sent two laser pulses through these “time slits”, in rapid succession. The position of the emitter, the mirror and the receiver remained the same – the only thing different was the time the lasers went through. Oddly enough, when two went through, an interference pattern happened. This was not an interference pattern in the same sense as with the

regular 3D space double slit experiment, though. This was an interference pattern that affected the laser’s frequency. Certain frequencies of light within the laser faded out, exactly in line with the way intensity faded out in the regular version of the double slit experiment. To visualise why this might be happening, let’s draw out this experiment in regards to time. The time slit experiment can be drawn in a similar way to the double slit experiment, except we’re going to need to visualise t

he change in the experiment over time. To do that, let’s create a 4D graph where space is along the x-axis and time is along the y-axis. This is easy enough to do – it just looks like this: The photon leaves the emitter to the left, hits the time slit, is reflected and arrives at the receiver. I’ve drawn this as a continuous line just to make things simpler later, but the idea works just as well either way. Later, a second photon is released from the emitter, it reflects and arrives at the recei

ver at a slightly later time, represented in how it takes place higher up (further into the future) on our time graph. If light behaved normally, travelling along at the speed it was supposed to go at, this would be the end of it. Instead, light is interfering with itself. This means it must be travelling along a path that takes it through the other slit as well as its own. This is the only way that light would come in with the pattern that we see. And just like in the double slit experiment, it

’s likely happening on the other side of the slits too: As for why it’s frequency and not intensity that’s being messed with here, think about the implications of what you might see if light did indeed come in at a different angle like this. Photons come in little packets of waves, as I’d previously mentioned. Now, look at what happens if you change the angle at which those waves arrive. Here’s how it normally might look: I’ve added a black timeline here, and have highlighted every time the rece

iver sees a new peak in the wave. Here’s what happens when you alter the direction of the wave’s arrival: Suddenly, the peaks are coming in much more frequently. The frequency of a wave over time is very much connected to the colour we perceive light to be. Low frequency light is redder in colour, while increasing the frequency shifts light’s colour towards blue. So, this colour variation makes sense. What makes less sense is what’s going on with the paths this light is taking through time. Reme

mber, the straight lines we started with represent the 299,792,458 m/s that we see light travelling. So what can we say about the photons that are travelling along these paths? For some parts of their journey, they are travelling slower than the speed of light, taking more time to arrive at a destination that’s the same distance away. And yet, for other parts of their journey, they are travelling faster than causality ought to allow. From their perspective, they are travelling backwards in time.

As a reminder, these two emitters on the left are actually the same one, just at different points in time. The same for the receivers on the right. It is a mind-bending result. And yet, according to the results of this experiment performed by a research team at the Imperial College in London, this is what is occurring. The implications of this are startling. Light always travels the path of least time – the route that allows it to arrive at its destination along the path closest to 299,792,458

m/s, the fastest anything in the universe apparently can go. And yet, it seems to me that in its efforts to locate exactly what path that might involve, light is testing the waters – putting out feelers that check to see if other paths, and seemingly other paths through time itself – might present a more viable solution. These feelers are interfering with photons that travel alongside it, but also with photons that travel a little ahead or behind it in time. To be clear, we never actually detect

photons taking any of these other paths. We don’t see photons coming in from the future. We never see photons travelling slower than the speed of light, provided there is no supercooling gases providing an explanation for why they slow down. And yet, for interference patterns to occur, to at least some extent light must be trying out alternative routes through time. Perhaps it’s like lightning, testing many different directions to find the optimal path to its destination, before finding the one

that works and collapsing down that path in one giant boom, all other feelers vanishing and collapsing: Or perhaps some other phenomena is at play. Who can say? For now, all we know is that light has proved once again that it doesn’t play by anyone’s rules. At least, not rules that we can figure out. Again, now might be a good time to pause, and reflect. This experiment we just saw hints that not everything in the physics world goes through time in the way we might expect. Light might be playin

g a little fast and loose with the linear nature of reality. We are comfortable with causality – with the idea of things happening one after the other, and things in the past influencing things in the future, rather than the other way around. This last experiment could be interpreted as throwing a bit of a spanner in that. But sadly for our aching minds, it’s not the only experiment to do so. Ok, break time is over. Can information travel backwards in time? It’s the sort of thing that would be r

eally useful, if it were true. You could tell your past self not to eat that burrito that didn’t agree with you, or could reveal to yourself the winning lottery numbers. But it just doesn’t happen; the resulting paradoxes alone make the whole thing laughable. In our universe, time always seems to flow in one direction – forward. The idea of travelling backwards in time, or even simply communicating with your past self, seems so outlandish, it can’t possibly be true. So, why is it that on the qua

ntum level, information seems to be doing just this? “Alex, stop!” you might be saying. “You’ve already shown us that the solid universe around us might be nothing more than probability waves, and that light has some weird element to it that causes it to interfere with other light in its past and its future. But this? Surely its impossible for information to travel backwards in time.” I understand the sentiment. It goes against all intuition, and by all accounts, it doesn’t seem possible. In pre

vious videos I’ve mentioned that objects would require infinite energy to even go fast enough to reach the speed of light. So how could something go so fast as to reverse the usual direction of time, and arrive at a destination not just instantly, but before they left? Not even light can do that, and it’s the fastest thing we know of. Well, this rule about causality’s speed limit seems to mostly apply to the macro-scale universe. And by macro-scale, I mean everything significantly larger than an

atom. But down on the quantum level, time might be obeying different rules… or at least, the speed of causality seems to come with some significant caveats. And to demonstrate this idea, we need to look at a man called John Stewart Bell, and quantum-entangled particles. I should apologise in advance for what I’m about to do to your understanding of causality. Ok, but what are quantum-entangled particles? In quantum physics, it’s possible to hit two particles together in such a way as to link th

em together, so that by measuring the one particle, you learn things about the other. For instance, if you know that the particles originally had a total of 0 momentum, and you learn the momentum of one of the new quantumly entangled particles, you know the momentum of the other particle will be the exact reverse – making sure that the total remained 0. Effectively, by measuring the one particle, you can learn things about the other. This works for other particle properties too, such as position

, polarisation, or spin. On the surface, there’s nothing too weird about this. It’s no different from me meeting up with a friend, and discussing our plans for the evening. We agree to go out, and we agree that I will pay for the evening and my friend won’t. Then, no matter how far we go on our night out, or even if we at some point separate, I know that I will be paying, and my friend will know that he won’t. This is how Einstein thought it worked. Only, it turned out that Einstein was wrong. B

ecause as it happens, me and my friend did not discuss in advance who would be paying. And strangest of all, we still both agree with each other anyway, 100% of the time, no matter how far apart we are. This is the strange thing about quantum entanglement, and quantum physics in general. We like to think of particles as having fixed properties. However, our penultimate mind-bending experiment shows that particles only have properties when you detect those properties. Yes, it’s like the double sl

it experiment again, only that was focusing on a photon’s position. It seems that particles are also kind of vague about the whole “properties” thing, instead only relying on probabilities, as defined by a quantum wave equation. This doesn’t make sense intuitively. Looking at a thing shouldn’t be what gives it properties… right? Well, how would you know? If a tree falls in the woods, does it make a sound? According to quantum physics, not necessarily. Let’s talk about the Bell experiment. The ma

ths for this is pretty complicated, but bear with me, it’s worth the ride. This experiment was instrumental in our modern day understanding of quantum physics, and closing off its loopholes earned Alain Aspect, John F. Clauser and Anton Zeilinger the nobel prize for physics in 2022. The experiment was first conceptualised by John Stewart Bell, who wanted to know if particles really did have secret properties that they carried around with them, known as hidden variables, or whether they really we

re making some of it up on the spot. He noticed an interesting mathematical fact about the spin of particles. Before we go any further, I should probably mention that quantum spin isn’t the same as normal spin. Misleadingly, quantum spin actually defines whether a particle is influenced – pushed or pulled – by a magnetic field. The name isn’t important, but it is important to note that these particles aren’t actually spinning, and so can have different “spin” values in almost any given direction

. Now let’s take two quantum-entangled particles, and let’s say that we’ve arranged it so that their spin adds up to a total of 0 between them. This means that if one particle would be pulled by a field, the other will be pushed by it an equal amount along that direction (with the understanding that this doesn’t tell you anything about their spin in other directions). One of the features of quantum spin is that if we measure an entangled particle’s spin in a given direction, let’s say up and dow

n, it will have a 50% chance to be spinning up, and an equal 50% chance to be spinning down. But remember, once you measure the other entangled particle, it will have a 100% chance to be spinning in the opposite direction to the first particle. On this fact alone, there’s no way to tell if the two particles already knew their spin, or are somehow deciding it on the spot and conferring with each other now that they’ve been asked. But Bell noticed a clever thing, by asking a clever question. If yo

u measured two quantum-entangled particles from two randomly selected directions, what are the odds that their spins for different directions would match? Let’s define that any time a particle is spinning towards a detector, its spin is “up”, and any time it is spinning away from a detector its spin is “down”. What are the odds that both particles would be spinning “up-up”, or “down-down” when tested, and what are the odds they would contrast? Let’s formalise this with a little experiment. Here,

we have two entangled particles, with three detectors reading their spin in different directions. If particle A and particle B are both read with the top detector, then one of their spins will be up and the other will be down. They are entangled, and this is what we looked at previously. However, if Particle A is read using the top detector, while particle B is read with one of the other two, these two directions of spin aren’t opposites, so Particle B has more flexibility in which way it goes.

Quantum physics claims the particles are making up their attributes on the spot, so once you’d measured the spin of particle A using the top detector, it was a 50:50 whether the spin on the other particle, using one of the other detectors, would match or contrast. But this is not what classical physics predicted. Let me show you what I mean. Classical physics claims that particles each carry around secret information defining their spin in any given direction. So, for our 3 tested directions, e

ach particle would have a value already. They aren’t making it up on the spot. Let’s say hypothetically our particles hidden information states “Up Up Down” for particle A, and “Down Down Up” for particle B, as B must be opposite to A for each of the directions 1, 2, and 3. Let’s pick out a random detector for A. We select detector 1. Detector 1 tells us that Particle A is spinning Up. Now let’s select a random detector for particle B. We select 1 there too. This detector gives us a reading of D

own. We can actually map out all the possible outcomes of this process of random selection in this graph. There are 9 possible outcomes if you were to only measure from two detectors at a given time: 1-1, 1-2, 1-3, 2-1, 2-2 and so on [2-3, 3-1, 3-2, and 3-3]. For each of these possible selections, we have fixed hidden variable results that we know already, because we hypothetically defined them earlier. Let’s fill them in now. Of course, if you test particles using the same detector on both part

icles, you’ll get a contrasting result because they’re entangled, but we’re not interested in these results. Classical physics and quantum physics both agree on this. So, let’s remove them. What are the odds that two different detectors for Particle A and B will see the same result, and what are the odds they’ll differ? Remember, quantum physics expected it to be 50:50. Particles are making up their values on the spot, and so it’s perfectly random which they’ll choose, as they aren’t confined by

the opposites rule. But in this table, classical physics says that contrasting results only happen a third of the time. The other times, they’re either both up, or both down. If we do this many times, assigning different directions each time, and ignore exceptions, for instance where the spins of the particles are all Up-Up-Up or Down-Down-Down - once you crunch the numbers, the important thing to take from all of this is that according to this maths, classical physics predicts a matching outco

me 55% of the time, while quantum physics continues to simply predict 50%, pretty table be damned. This percentage difference was key. By quantumly entangling particles, and running this test over and over again, you could now see which percentage was correct. And it turned out the winner was quantum physics. Particles were just apparently making up their spin results on the spot. Which is spooky. Because not only does that call into question our perceptions of reality itself, but that also mean

s that the moment one particle decided on its spin result, its quantum-entangled partner instantly knew that that decision had happened. You could test both particles at once, no matter the distance, and this same result would come back. Somehow information had travelled from the one particle to the other in no time at all, far faster than light itself. So already something strange was going on here. This result disproved Einstein’s predictions, and showed that some information does seem to go f

aster than light. But we can take this one step further, and have information going back in time. There is another experiment , our last experiment, known as the “delayed choice” test. Its primary purpose was to explore the fundamental nature of light – whether it was a wave, or a particle, and to figure out when it decided to be one or the other. Unlike the double slit experiment, though, this test was more about that last part – trying to identify the moment the waveform collapsed down into so

mething discrete. In the double slit experiment, light seemed to choose a different path through space depending on whether it was observed, or if it wasn’t. In 2006 a number of scientists asked an interesting question: what would happen if you tried to observe the light after it had to pick a path? Consider the experiment: A single photon is sent into a Beam Splitter, with a 50/50 chance of either being allowed to carry on its way along path 1, or getting reflected up along path 2. Once on eith

er path, the photon is bounced off mirrors, with both paths reconverging here, where another beam splitter is inserted. Once again, the photon has a 50/50 chance to go either way, with an even chance of arriving at one of the two detectors. If light were just a particle, sending a single photon into this experiment would give you an even chance of it arriving at one detector or the other. You’d not be able to tell which way it went, as the two beam splitters make that impossible to know, but you

could see where it ended up. However, this does not occur. When the second beam splitter is present, the light produces an interference pattern, indicating that the single photon went down both paths, ultimately bumping into itself, before moving on to both detectors. This seems like strong evidence that light is a wave; it certainly behaves like one here. But what happens if you remove the second beam splitter? Suddenly you know which path the light travelled down – if light arrives at the top

detector, it must have arrived from path 1. If it arrives at the side detector, it must have come along path two. And something about this knowledge spooks the light. It stops going down both paths, and suddenly each photon only arrives at one detector. Here’s the question – what happens if you insert the beam splitter after the photon has already started down either one or both routes? This is why the test is called “delayed choice”. If you delay choosing how exactly you intend to detect the p

hoton, whether by knowing which path it came down or making that ambiguous to you, what happens to the light? What happens is a very strange thing. When this experiment was performed, it was done multiple times, with the beam splitter randomly being inserted or not, but always being inserted after the photon had entered one or both paths. And yet, the results came back unequivocal. If the beam splitter was present, the photon suddenly, and seemingly retroactively, stopped picking a path. If the

beam splitter was removed, the photon seemingly knew it would later be detected and picked a specific path to accommodate. Somehow, the beam splitter being added or removed in the future changed what the photon did in the past. So, what is happening here? Is it really true that particles somehow saw the future? Did the experiment cause information to be sent back into the past? Or is there some other principle at play here that explains this whole thing; that accounts for the instant transmissio

n of information between quantum particles, and allows it to be perfectly rational that light could travel down one path or both at the same time. Personally, I’m inclined to think that this is more likely. We clearly don’t understand what is happening here. But it must be admitted; if we don’t understand what is happening, there’s nothing saying that causality isn’t being ignored. In some way, maybe on the quantum level time really is more fluid than it is up here in the larger universe. Maybe

space and time simply do not apply down there. And maybe one day someone will be able to come up with a theory that allows all these strange phenomena to make finally make sense. Until then, we’ll just have to keep asking the same question: Can information travel backwards in time? Until then, we’ll just have to all agree on one thing: quantum physics is strange.