Eyeballs are funny things


This optical illusion is making the rounds. There are twelve black dots located at some of the vertices. Can you seem them all?< ?p>

opticalillusion

You can’t see them all at once, at least: focus on one, and all the others disappear. This is not at all surprising — the central area of maximum resolution in your eye is very tiny, and your peripheral vision simply isn’t very good.

What did intrigue me, though, is that if I focus on a central intersection I can see two dots at once, one to either side of the intersection I’m looking at. However, I can’t simultaneously see two dots on the vertical plane.

I think this means I’d be particularly susceptible to drop bear attacks.

Comments

  1. alkisvonidas says

    However, I can’t simultaneously see two dots on the vertical plane.

    I can. And I think I can even see 3 dots in a row on the horizontal play, but possibly my eyes are darting a bit while I do so.

    I can also rapidly scan a row and experience 4 dots at roughly the same time.

  2. bittys says

    I can see two in the vertical plane (and the ones at all four corners of a square if I focus on the middle point).

    Question: Is this an age related thing, or just general human variance?

  3. Marshall says

    What’s crazy about this illusion is that you can train yourself to see more. Here’s a technique which is still a bit difficult:

    1. Only consider the bottom row of 4 points. Stare direction into the center concentrating on viewing the outer two dots. With a lot of practice and concentration, they’ll eventually come into view.

    2. Move to the middle row, do the same. Once you have that, try to include the bottom row.

    3. Move to the top row. Move back to middle, and try to include both top and bottom.

    What’s interesting is how this demonstrates some of the complex interaction between conscious control of attention and the automatic earlier stages visual processing.

  4. chigau (違う) says

    I can see two at the same time both horizontally and vertically.
    and occasionally, four in a square

  5. blf says

    The mildly deranged penguin has a suggestion on how to see all the dots at once: Imagine they are ninja peas coming out of the lattice-work to attack. Now you will see them all, out of sheer terror if nothing else… (then, with you now cowering behind your overturned desk, she runs off to eat your cheeseboard).

      ────────────────────────

    I think this means I’d be particularly susceptible to drop bear attacks.

    Sharply-pointed Wizzard hat suggested.

  6. andyo says

    Yeah, it depends on how big the pic is on your screen. I can see more dots when I make it bigger (counterintuitively, I guess). Just try zooming in the pic in the post, see the difference instantly.

  7. slithey tove (twas brillig (stevem)) says

    Can you seem them all? simultaneously ???
    FTFY
    the question, when I first encountered this illusion was specific about asking if one could see all 12 at once.
    Yes it is possible, easy, to see all of them, each one individually. Yet to see all 12 at once is quite uncommon. Not necessarily impossible, just requires an extraordinary size of the central region of the retina.

    yes vision is weird, how the processor fills in gaps with ‘best guess’ approximations.

  8. frog says

    I can see four dots on the corners of a square if I stare at the intersection at the center of the square.

    Also fun: moving your eyes rapidly all over the figure and kinda-sorta seeing most of the dots at once thanks to lingering afterimages.

  9. anbheal says

    The drop bear attack bears a bearing on this issue. I think. I was at a cocktail party a couple of years back, where Roy Glauber was the father of the hostess, He won a Nobel for inventing quantum optics in the early 60s. He was also the youngest member of The Manhattan Project. Smart dude, but not a biologist. Anyway, he overheard me mentioning to some people I was chatting with that I could see the individual blades of a ceiling fan when I gave them a sideways glance, but not when I looked straight at them. He sidled over and muttered quietly, “I think it might be because we were never meant to be predators, and so are wired as prey”.

    I thought about this for a moment, and then nodded. “You’re saying that hunters scope to the front, the hunted scope to the side?”

    “Maybe”, he replied. “Physicists don’t make up QUITE as many stories about the past as biologists do, but it seems as though it could explain the phenomenon.”

    Anyway, I see them all when I give them a sideways glance.

  10. says

    Nope, I still can’t see them all at once, and I’ve got a distinct spatial bias for horizontal vs. vertical.

    re #12, this clearly means that I have a predator’s eyes, unlike the rest of you sheep. Although I’m not interested in hunting birds, just you mammals arrayed on the savannah before me.

  11. Marshall says

    PZ @14 see if you still retain that bias with one eye closed–the horizontal bias is probably because we have two eyes situated horizontally.

  12. martin50 says

    PZ–re: fixating on a central intersection and seeing both–my initial hypothesis was that it was a result of our eyes being on a horizontal axis. However, when I cover one eye, I can still clearly and persistently see two horizontal dots when I fixate on a central intersection. (The same thing is initially true vertically but they fade very quickly to nothing.) Turning my head rotates the effect with my head (i.e. it’s my head, not the grid.)

  13. Becca Stareyes says

    I can see multiple dots if I keep my eyes moving across the field. A bit like what I have to do to see a ‘faint fuzzy’ at the telescope: try not to stare at one point directly. (Since in that case, the cone cells in the center of my eye are less useful than the rod cells that are concentrated in the periphery.)

    But still, it’s so neat to watch how my visual processing hides some of the dots based on the surrounding pattern.

    * That is, any extended faint object, like a nebula, galaxy, or globular cluster.

  14. Marshall says

    @martin50 #16 – Our monocular field of vision is a bit larger horizontally (approx 160 deg) than vertically (approx 130 deg). Using both eyes, it’s expanded to about 200-220 deg horizontally.

  15. Rich Woods says

    However, I can’t simultaneously see two dots on the vertical plane.

    Turn your head through 90 degrees. Works for me, for a short time anyway.

    No, not that axis, the other axis. One of the others. If you’ve got more than three I can’t help you.

  16. says

    I can see two horizontaly and verticaly when looking between them, and four if I look at the middle of the square they form. I tried this with both eyes and with one eye only.
    I have noticed that distance plays role, as well as much I focus -the more I focus on the intersection, the worse I see the black dots, and when I focus very intensively, they disappear. So I guess that seeing more dots at once is due to eyball wobbling and the wetware compensating.

  17. komarov says

    Focussing on the centre I see the two adjacent horizontal dots. Fading in and out, out of phase. Drat!

    It’s either a psionic signal from the aliens or minute movements of my eye as it tries to see both dots while pretending to stay focussed. Given my legendary visual acuity (I was awarded many contact lenses for it) I’m going to go with aliens. Explains the splitting headache I have, too, so take that for evidence, skeptics!

    P.S.: Turning my head sideways* as suggested by Rich Woods (#19) I can see both dots at the same time. But it’s stll the horizontal ones when I would have expected the vertical ones. I guess my eyes are more efficient when rotated by 90°. Perhaps they’ll get better still if I stand on my head…

    *On what I shall call the nose axis, this being the most prominent identifying feature on it by far. The other axes are the nostril axis and the ear canal. There, nomenclature sorted.

  18. says

    Yeah, must be very dependent on screen size. Mine is.. 23″ wide screen and if I focus center I can see the two center, and sort of see the ones above and to the sides, but not the corners. Focusing on a single point I can, again, sort of see six.. the one I am focused on, the ones above and below, and the three either to the left of right of the focus. But, they are like fuzzy black blobs, instead of dots.

  19. Azkyroth, B*Cos[F(u)]==Y says

    I find I see one row at a time “by default,” all four dots at once, unless I focus on the specific one furthest to the outside. I’ve managed to get two on a vertical plane to appear at once, repeatedly, and a four-dot “T” at least once…

  20. Azkyroth, B*Cos[F(u)]==Y says

    With one eye closed I have to squint and wiggle my eye to see four in a row. Still can’t get a column of 3.

  21. Terska says

    I though sitting further away from my laptop screen might allow me to see more dots but it doesn’t help me at all. I can just barely see four at a time in my peripheral vision.

  22. unclefrogy says

    that is a cool demonstration of how we actually look at things. Our eyes and the visual system does not really work like a camera does. We continuously scan around and not just look out at the direction we are facing.

    it is rather fast we do it all the time and do not notice we are doing it this “illusion” highlights it and what others have described are hints on how to ”defeat” the scanning.
    I find it much easier to look at a “normal” style painting even a purely abstract one than a photo-realist painting and longer than most photos. The focus in paintings is not uniform and meshes with the natural scanning the visual system uses instead of forcing the blank stare like a photo or photo-realist.
    All that art and painting stuff is subjective on my part of course. the scanning is real though is real
    uncle frogy

  23. Crimson Clupeidae says

    This is also a good demonstration of why you shouldn’t fixate on a point when driving. It’s easy to miss smaller objects like motorcycles and bicycles.

  24. Tethys says

    Huh, it is interesting which dots disappear. I lose the dots immediately above and below the focus, and to the left horizontally, but all the others are still there though slightly less in focus. It makes my eyes feel twitchy to look at it for very long.

  25. Mobius says

    I can see two dots on the horizontal or vertical if I look at a point in between them. Staring at the center of a square, I can see all four but they are a bit hazy, as if I am seeing things.

    Slowly scrolling the screen, all twelve jump out at me. I think this is likely an effect of movement in our peripheral vision bringing things to our attention.

  26. FossilFishy (NOBODY, and proud of it!) says

    Crimson Clupeidae #27:

    This is also a good demonstration of why you shouldn’t fixate on a point when driving. It’s easy to miss smaller objects like motorcycles and bicycles.

    Yes, looking at a single point when driving can be a problem, but so too can sweeping your eyes back and forth.

    There’s a phenomenon called saccadic masking that happens when you sweep your eyes. The image you see as your eyes are moving is a construct of your brain. You only really see at the end points of your sweep. Magicians use this to hide moves from the audience and fighter pilots are taught to scan in small increments. That method: looking at one point then, another a little farther along, and another, and so on is apparently the best way to avoid this.

    Mind you, I’m no expert, and I wonder if there is a slow enough steady sweep that will avoid this masking.