This wonderful illusion was created by Frederick Kingdom and colleagues at McGill University, and is one of my favourites. Big Ben appears to tilt more in the photo on the right, but in reality the two photographs are identical….
Any idea why it works?
Richard Wiseman 
I’m going to guess that when we look at the image on the right, the left image’s clouds are processed as well. So our brain thinks there is more room on the left than the right, and crams it all in from the left, making the image seem tilted left.
You’ll notice that Westminister itself doesn’t appear tilted, because the other side (following my theory) doesn’t have clouds, thus not resulting in an illusion of space.
What would happen if one placed three of the photos side by side?
Quite frankly (and I say this without having read the comments), you have got me on this one!
I think your brain supposes the perspective of the left photograph continues across the images to the right, and then the big ben would lean very much to the right
My guess is, it’s our brain trying to project proper perspective onto the second tower – if the two towers were actually right next to each other, the only way you could get them to look the way they do here is if the second one was tilted.
Not sure I explained that well, but it’s what I got.
This has more Verisimilitude than any other explanation.
I have this impression that @Naranda and @Robert Irving III, presented the same explanation, almost at the same minute too. I agree with both. It is not only more tilted: the right tower appears to me to be slightly bigger (which it plays well with the explanation given), and strangely, a little more lighter in color. If I interchange the photos, the right tower becomes bigger and tilted. If I add a third equal photo, the rightest tower becomes the most tilted one, after the second. With three towers, there is a strange impression that they irradiate from the same center. Note that if two towers are perceived relatively tilted, they’ll appear to cross of course, but that doesn’t tell us if they cross always at the same point. With 3, that becomes apparent. Considering that the standard way to create a perspective in drawing is to put parallel lines coming from a common point, yet, we interpret them as parallel, this suggests that the brain is trying to correct the perspective effects, and doing so, they conclude the towers are not parallel.
I also did simple drawings but failed to have the same response. To “believe” the images to be representing 3D objects must be essential.
the right side of the tower in the photo is nearly parallel with the right edge of the photo, making the left side of the tower look more tilted in comparison due to the upward perspective of the shot.
My guess is that our brain processes the angle from “vertical” for the big ben on the right as being relative to the big ben on the left, so we think it’s leaning more.
I would guess that you are comparing the angle of the right wall on the left picture and the left wall on the right picture and this leads you to thinking that the right is leaning more.
I concur!
What Gillian said. I wish I had said it first.
yep I agree…… I down loaded and printed then put one on top of the other and held it up to the light and they are the same except the right photo looks like the bottom was cropped just a very slight bit long….as Alex noted…. I trimmed that off….still had the same tilted effect…. so I think Gillian got it right…..Love the name “Gillian”!
didn’t i say that first?
well we all sort of think about the same
I’m with Gillian.
What impresses me most is that watching a single image i’m not VERY struck by the weird angles.
It’s the perspective, I’d guess. Our eye wants to put the tower in the right picture on the same plane as the one in the left. Because they have no real spatial relationship with each other and the tops of the towers are further apart than their bases, the right tower appears to be leaning away from the other
EASY! The left side of Big Ben (in this picture) slopes more than the right side, so when putting duplicate images together these get compared creating a noticeable difference in gradient. Easily proved if you place your 3 middle fingers in the gap between, this makes them look parallel again.
Because our brains are dumbasses š
the left side of the tower is more tilted than the right side. the right picture of the tower looks more tilted than the left because putting a diagonal line beside a straight line will make the diagonal line more slanted.
Maybe it’s the vertical green line.
It doesn’t work for me. I saw it before reading the post below – immediately I knew it would be something along the lines of, “they’re the same, but they look like they’re different”. But I expected something else. They just look like they’re the same to me.
In other words, I declare myself optical illusion-proof.
It spears as though the angles and spaces (sky) of the.first image combine with and influence the appearance of the 2nd image – I guess the effect would be reduced with an increase in space between the two images, or indeed increased to an extent with repetitions of the image
What happens if you swap the two pictures around?
Good illusion – so strong an effect I’m having to take your word for it that they are the same!
As others have suggested, I’m supposing the brains expectation of perspective is the cause. We know our brains lie to us about what we see all the time by making things up it thinks should be, and our whole perception of ‘leaning’ is artificial given it’s actually a flat 2d image anyhow so one could say even with a single image of big Ben that the perception of it being a vertical structure is an optical illusion – it’s just one that we’re familiar with and reflects what we expect of the world š
look at the very bottom of both images, you will find that there are some things that appear on the left image relatively more than they appear on the right image. which means that the image (not Big ben) on the right side is tilted a bit towards the right and then cut in a straight square shape. So the building appears tilted.
I mean “things that appear on the right image relatively more than they appear on the left one”.
There not quite identical… you can see the lower left side is different. so i beleive one is slightly higher then the other. the angle looks the same. compared them by super imposing them in photoshop with a transparency. there not quite identical but i dunno if it changes anything. the angles seemed to be the same.
The tower on the left creates a mental border, which we use as a reference point (subconsciously presumed straight), and as a result see the tower on the right angled with respect to the first.
I put the two photos over each other both left and right and both photos a exactly the same photos…they line up perfectly…..so I guess it would be due to the stright line of the photo edges between the two that trick your eyes/brain. Either way cool!
I think it’s because the slope on the left side of the clock is larger than the slope on the right side of the clock. So when you look at the middle of the image your brain compares the two slopes and comes to the conclusion that the right image has more tilted clock.
Less sky on the right of each picture, so the one TO the right appears to be filling in the gap with more sky on the left.
It’s been said before, but your brain expects slight differences between left and right images – as this difference is not there, it confuses the issue by fooling us into thinking the images must be different, and hence the slight fall off.
Interesting thing is that if you cross your eyes, then slowly relax them, you will start to see a third image in the middle. Concentrate on that, and you’ll see a “slightly flattened” 3D image.
You can see lots of examples of this technique here: http://www.flickr.com/groups/stereographicshowcase/pool/
when I look at it with my eyes crossed and see the three……they then look all alike at the same angle!
Previous comments are very plausible – the brain looking at the right side of the left tower cf the left side of the right tower and making assumptions about what is vertical: everything else becomes relative to that.
Interestingly, if you take a cross-eyed 3D stereogram look at the pictures (so that you see three images – one central and two peripheral) the 3rd image does, indeed, lean even more. How many images would you need in a row to have a horizontal tower…?
They are not identical, the left one is moved slightly down and right.
slw print them, cut them, over lay them….then see how they look…. it’s cool! …… :}
I thought so too as you can see a few bricks in the bottom left corner in the right one and not in the left one
It’s all to do with all to do with the way a wide angle lens on a camera causes the verticals to converge to an imaginary focus point. The right image looks more tilted relative to the left image which has horribly converging verticals. So your brain tries to compensate for the converging verticals and makes the illusion happen.
The only way to avoid that is with an architectural camera where you can shift the lens with the bellows. Or by taking the photo from a higher point so that the verticals converge less.
http://photonotes.org/cgi-bin/photo-entry.pl?id=Convergingverticals
As some has suggested – its perspective.
You expect the towers to converge as all tall structures would when you look up at them. The expectation is strengthened by the fact the parallel sides of the towers converge with distance. Because the towers DON’T converge you think this is because one is leaning away from the other.
Because the angle between the two towers is the sum of the angle on both sides (A+B). While the tower on the left is tilted only A degrees. We rely on the edge of the tower on the left as the guideline for the tower on the right. I had a lesson about this in graphic design.
because the right one is a little bit moved up and right from the original center of the left one. you can see for a prove of that at the bottom left corner of the right picture there are some objects not appearing on the left one and also the top of the tower on the right is closer to the end of the picture than on the left one.
and that difference is playing an eye trick so you can assumed that the right picture is different from the left one.
Something to do with me absolutely loving grape jelly.
good illusion. and @J, i think it is NOT RELATED WHATSOEVER to grape jelly. maybe strawberry jelly, but DEFINITELY not grape flavour.
strawberry JAM…..not jelly……giggle or maybe apple butter? …..humm…
I originally thought it was comparison to the left wall, but after reading the comments I concur that it is that the angle of the vertical lines within the picture gradually slant across left to right.
What is interesting to me is that we are all seeing it as the right tower falling over.
Is there a way the picture could be taken that would make the left tower look like it was falling the other way, or is there something about the fact that we are a “left to right” culture that makes this work?
My guess is that, because the edges of the tower are perceived are parallel (though, in the image, they are not), the fact that the left side of the right tower slopes further than the right edge of the left tower, it is perceived that the right tower itself is sloped further.
It definitely seems to be a result of the perspective. Robert Irving III, said it well up top, but there may be another reason: that the 2 edges of the Big Bens that are closest to each other diverge.
Haven’t looked at the other comments but it seems to be an issue of which visual reference line you pick up on: (1) the other (leaning) tower, or (2) the vertical line between photos. Use (1) & it looks as if the second tower is leaning more, or use (2) and they are leaning the same amount ??!!
The image on the right looks like the brightness is turned down to me.
your mind processes them as one picture, not two, and so the second “half” of it seems skewed since if it WERE one picture and the second half were a continuation of the building, it WOULD be tilted.
Sorry: but they both look tilted.
The tower has two edges. the one on the left and right of the tower are not parallel. we compare the left of the right tower’s edge to the right of the left tower’s edge.
(if you compare the “right” edges, it looks parallel)
The thick, central border draws attention to different halves of the two pictures. Ah’ guesseses.
Easy peazy. The photo on the right looks more tilted because all three of the “vertical lines’ in the tower are not parallel in the original picture. From the perspective of the camera, the lines of left-most edge on the clock tower and its 180 degree opposite corner would converge if drawn out. When the pictures are placed next to each other, the supposedly parallel lines become even more noticeably not parallel, and we attribute this to tilt rather than perspective.
It’s a ‘cross-eyed’ 3d pic.
Sorry Joe,
But when I do the X-eyed 3D thing it still looks flat and tilted.
Unlike the ones in my collection this one doesn’t pop out or into the screen.
I think the two are essentially the same image.
I did a screen capture to both photo & use a photo manipulation software to swap the photos (left to right & right to left). The result is still the same, it appear that the Big Ben still appears to tilt more in the photo on the right.
Interesting, I believe it is due to the edge of the building, the left edge of the building is more tilt than the right edge. When two identical photos are put together, our eyes focus on the center (which is the right side of the left photo & the left side of the right photo) making us think it in such a way.
Tried using stereo-vision on the photo (cross eyes slightly and focus on the image created by both images). Both are sure identical.
If you look at only one of them at a time, they seem identical. So it has to do with the image as a whole
The two images aren’t quite the same. There’s a bit on the right image that’s cropped out on the left one.
That’s not my point though. My point is, cross your eyes until the 2 images merge, and then sway like a drunk man. Big Ben goes a bit 3D! It’s a miracle!
The images are identical! Print it out if you have to, but they are the same!
Sorry, I take it back, after printing they are not identical, the cropping is apparent now, though I’m not sure if that’s what is causing the illusion.
These two images are not identical. Look very carefully at the bottom, you will notice that the “poles or beams” on the left are cropped more than the one on the right. Therefore, the right image is leaning towards the right!
no idea what it means but in a recent college humor article you can see the same effect if you look at any pair of church signs (identical other than the words)
sorry i didn’t put a link. http://www.collegehumor.com/article:1801179
Although I arrived at the reasons independently, I have the same reasons as suggested by Dougie. It is a distortion caused by the slightly wider angle of the lense for the image on the right.
I believe that the illusion happens because in the first image you see Big Ben in relation to the upright of the side of the images frame. However when you look at the second image you look at Big Ben in relation to the first Big Ben and not the vertical line in the middle. Because you look at it like this you add the all ready angled look of big ben (a) with another of the same angle thus making it look like its twice as angled as the first image.
The vertical cropping of the building on the right side of lefthand photo makes the angular tilt of the left side of the building in the right photo more pronounced. The eye ‘extrapolates’ this optical anomoly upward along the height of the tower making the entire picture appear more tilted.
its our brain that assumes the first big ben as the Vertical axis, making the secong look more tilted… gud one for sure…
first i tried to observe the background , to see if the
a second look got me this… THE IMAGES ARE NOT IDENTICAL !!!
the second is tilted in its horizontal axis for sure… and then the background clouds are IDENTICAl to the first one……..
MASTERpiece!!! but with a small glitch…. the second image shows a small view of the building thats not in the first one…. have a keen look at the bottom left corner of the 2 images….!!!!
Your mind tries to resolve the “inner” corners (right corner of the left pic and left corner of the right) of the right and left pictures as being parallel, but they aren’t! The perspective given by the angle of the photograph creates am upward taper to the pictures that prevents them from being visually parallel causing the illusion.
The right picture is a teensy-bit higher than the one of the left..?
well, if you see the bottom, you’ll see it.
but i don’t know if that’s the reason….. aha.
My partner and i wouldnāt intellect to become a article or even elaborating with many of the themes you generate relating to here. Yet again, wonderful web site!