Cognitive Daily

Casual Fridays: Veiled illusions

Last week’s Casual Friday study was all about illusion. For example, you may have thought our goal was to see how well you could recognize an illusion. However, we really just wanted to know what kind of computers our readers use:

i-4edcd4585eadf253354a16795c419d5a-illusions1.gif

Amazingly, Cognitive Daily readers use Macs at a rate (22.8 percent) about seven times higher than the U.S. market share of Apple Computers (roughly 3 percent).

We did also want to know something about how you see illusions, so we designed a simple experiment based on a brilliant illusion by Akiyoshi Kitaoka. If you haven’t visited his web site full of astonishing illusions, why not head over there now? We’ll still be here in a couple hours, when you’re finished.

Our experiment was based on this illusion:

i-9599cf538936b1ee7f5b60e6ba86e079-illusions2.gif

When most people view this figure, which is in fact a static image, it appears as if the diamonds in the outer rings are slowly moving clockwise (to the right). The diamonds in the inner rings seem to be moving counter-clockwise (to the left). But if you view an extremely faded version of the same image, you almost certainly won’t see any motion at all:

i-dea498f7135ec03a81827a08dde788ba-illusions3.gif

For our experiment, we divided participants into four roughly equal groups by asking for their birth month. One group saw the original, unfaded illusion, and the others saw progressively more faded versions of the same picture, with the final group seeing the most faded version we showed you above. Here’s what viewers said they saw:

i-3dbda45648c964ec0f235baec216b887-illusions4.gif

As you might expect, viewers saw progressively less motion as the illusion faded. Impressively, even at the highest level of fading, some viewers saw the motion and correctly identified the direction of the motion. I had predicted that there might be a difference between Mac and Windows users, since often the same image appears darker on a Windows machine, however, we could find no significant difference between computing platforms.

We did notice one startling finding based on handedness. None of the left-handed users made an error of direction of motion:

i-e84497e6bd8f07cf4958c7e1f735723f-illusions5.gif

When the right-handers’ “can’t remember” responses are combined with incorrect motion direction responses, the difference is significant. Is there some difference in the visual system of left-handers that explains this? Are left handers more careful than right handers? And why are so many Cognitive Daily readers Mac users? We invite your speculation in the comments.

Comments

  1. #1 Carl Manaster
    February 17, 2006

    What are the numbers of right- and left-handed respondents? I suspect that the “startling finding” is just the result of a too-small N.

  2. #2 Matt McIrvin
    February 17, 2006

    The market share of Apple among home users is probably significantly higher than 3%. It’s that low overall because of office IT purchasing.

    22.8% is anomalously high, though, for a site that isn’t Mac-specific.

  3. #3 Tim
    February 17, 2006

    Mac’s are pretty prevalent in academia. Thats why I have one.

  4. #4 jepalmer
    February 19, 2006

    To add to the question about your N, did you get a higher-than-expected percentage of lefthanded participants? And what was the lefthanders’ rate of using Macs? Just curious. . . as I fit both categories. :)

    BTW, they look like little red diamond-shaped planaria. Adorable!

  5. #5 Dave Munger
    February 19, 2006

    Carl,

    There were 21 left-handers and 195 right-handers. We had a total of 250 responses, but some didn’t complete the survey, and I filtered out all responses with noncorrectable vision impairments. But we’re really talking about 42 cases here, since respondents had opportunities to make errors in the direction of motion for both the inner and outer circles. I wonder if left-handers are just more conscious of left- and right-direction: since things like screw-tops and scissors are designed for right-handers, left-handers must make active decisions about left- and right-handed motion more often.

    Jepalmer, there was no difference in computing platform use among right- and left-handers. However, the popular Mac fan site MacRumors surveyed its users a few years back and found a much higher incidence of left-handed Mac users than right-handed users than in the general population.

    I suspect this result might be due to selection bias: more left-handers will respond to a survey about handedness than right-handers, since handedness is a much more relevant issue to left-handers.

  6. #6 Katherine Moore
    February 19, 2006

    When calculating the right/left handed difference, did you factor out the effect of fading? You want to make sure there is an equal number of lefties and righties in each condition, and with a small N like that I could see how there could be a bunch of lefties born in June instead of January, for example. If the left handers saw the least faded items in higher proportion, it would make sense that they wouldn’t be making as many errors.

    As for the Mac users phenomenon, I’m not in the least bit surprised. For whatever reason, academics and academic institutions tend towards using Macs. I’m sure there’s a greater percentage of cognitive daily readers who are academics than there are academics in the general population of computer users.

  7. #7 Katherine Moore
    February 19, 2006

    Scratch that thought — I just noticed that people made as many errors in the “no fade” condition as in other conditions. In any case, I still think that it’s hard to compare right and left handers when there are so few left handers.

  8. #8 Dave Munger
    February 20, 2006

    I agree, Katherine, there could be some other factors at play. We’ve got sampling issues, platform issues, all kinds of other factors that could influence this result (it’s casual, remember!). But I’ve yet to come up with a factor that could explain the significant difference between left and right handers. There are plenty of studies that use 21 total participants (and plenty more that use fewer), so an N of 21 in itself isn’t enough to explain away the significant difference we observed.

    So until someone can come up with something better, I’m sticking with my original hypothesis that left-handers are simply more attentive to left-versus-right distinctions than right-handers.

  9. #9 Dave Munger
    February 20, 2006

    Oh, I should also add that left-handers were indeed distributed evenly across the different fading conditions.

  10. #10 Cat
    February 20, 2006

    I can’t remember, but did the survey track if participants were male or female? Males are supposed to have slightly better visual-spatial processing — it’s possible that the left-handers were mostly males.

    Or, it could be due to lateralization. Visual-spatial abilities live in the right hemisphere. Since the right hemisphere controls the left hand, it could be that left-handedness provides an overall advantage for right-hemisphere processing.

    Of course, in both cases, you’d have to assume that visual spatial *memory* is similar to the the actual processing. I don’t know much about lateralization & sex differences in spatial processing, but maybe another reader knows more about this?

  11. #11 joseph rainmound
    February 22, 2006

    that old chestnut? the market share indicates the numbers of computers sold; the number of people who own a specific type of computer is entirely different, as Jobs well knows. Meanwhile, on the other end of the spectrum Apple computers tend to last longer (my old iMac is still going strong) so people need to replace them less. So we have a case of people buying Apples less than people need to buy Windows machines.

    Say Bob buys a powerbook. Chances are this will be good for 3-5 years, with upgrades. In the same amount of time, a businessman might go through three or more laptops in my agency. So the market share of PCs vs. Apples is 75%; the population spread is 50-50, and the businessman is broke. (The apple user is possibly an artist and therefore broke anyway.)

  12. #12 Paul D
    February 24, 2006

    Your readership’s computer statistics are interesting. For starters, for reasons cited here and elsewhere, the actual percentage of computers in use in the US is probably more like 5%. Then consider the higher percentage of students and professors on Macs, the greater level of computer proficiency and trend-setting among Mac users, and thus the overall higher percentage of Mac users among blog readers and feed junkies.

    40% of the visitors to my own language-related blog use Macs, probably due to all those reasons plus the ease of using Japanese on a Mac compared to Windows.

  13. #13 Carol
    February 25, 2006

    I seem to recall you asked who was using laptops. If all your errors were from people using laptops (or flat screens) I submit that is more likely the source of your errors than handedness. I’m using a laptop, and in order to see anything at all, I had to bend my screen backwards and forwards. I was quite surprised, when I finally saw the actual image, what the colors were and where the dots were. I could hardly see them no matter what I did.

  14. #14 gen
    July 20, 2007

    make more!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Nice… bravo….

  15. #15 mandi
    June 2, 2008

    i love illusions- they r so cool- espissally animal ones- l8er

  16. #16 laurie
    June 3, 2008

    I had forgotten this Illusion. Thankyou!

  17. #17 amy phillips
    September 9, 2008

    MAke more illusions for this site