Waste time at work for science!

Help out some researchers at Harvard by taking this 5-minute test on short-term visual memory!

I scored a 3, which is slightly above the supposed average of 2. It’s interesting because I’ve always had a problem with one specific kind of visual memory — faces. I can’t remember a face (short-term or long-term) to save my life, but apparently if all the people I meet every day had faces composed for an Atari game, I’d be okay.

Rebecca Watson

Rebecca is a writer, speaker, YouTube personality, and unrepentant science nerd. In addition to founding and continuing to run Skepchick, she hosts Quiz-o-Tron, a monthly science-themed quiz show and podcast that pits comedians against nerds. There is an asteroid named in her honor. Twitter @rebeccawatson Mastodon Instagram @actuallyrebeccawatson TikTok @actuallyrebeccawatson YouTube @rebeccawatson BlueSky

Related Articles


  1. I only got 2.5 – and I wasn't able to say 'cat' or 'dog' to myself over and over during the whole test. I'm not even conscious of when I stopped…

  2. 3.3 for me. I had to guess several times, which turned out to be correct roughly half the time, I'd say. I also noticed sometimes I could describe one or more shapes to myself, which helped a lot. Only visual would have probably brought me down some more. Though I wonder what the highest number would be – 4? 5?

  3. Why does it let you specify your age as 1 years old when the test minimum age is 18?

    I have a really hard time recognizing faces in real life – show me two pictures of the same person (along with other people that aren't the same between the two pictures) and I usually can't tell you who the common person is. It's really a problem when watching movies, as I often get characters horribly confused…

    I thought I was horribly screwed on the practice run – I got every one of those wrong. On the actual trial I got a 3.5959595959596 (exact number provided). That's way more precision than I think is justified…

    I don't know if that means I'm good or bad? I only remember missing two or three in the actual trial.

  4. Rebecca & ydant,

    If you really have problems with faces, you should email Ken Nakayama at the Harvard psych department. He recently discovered that there are a lot more people with face-perception difficulties than anybody knew. Anyway, he is always looking for new people who would be interested in participating in his experiments — some of which are also online.

    or you can email me ([email protected]) and I'll put you in touch with somebody from his lab.


    Responding to some other comments…

    The top score in this particular experiment would be a 4, which would mean you missed nothing. Some of you are doing amazingly well.

    I don't want to post here the reasons we allow you to put in an age under 18, but if you email me, I'll explain it.

    Thank you all!

  5. Berandor said,

    I noticed my “dogging” of the questions started to lag several times, as well, and had to consciously remind myself to keep saying it.

    Yes, I had real trouble remembering to think the word, especially near the end.

    ydant said,

    I thought I was horribly screwed on the practice run – I got every one of those wrong. On the actual trial I got a 3.5959595959596 (exact number provided). That’s way more precision than I think is justified…

    Now that's very interesting, because I did terribly on the practice run, as well. I even thought to myself, "Man, these things all look the same. Like faces!"

    vacognition said,

    If you really have problems with faces, you should email Ken Nakayama at the Harvard psych department.

    Ooh, I might! I've heard of a disorder in which people just can't remember faces at all, even the faces of long-time friends or relatives. I'm not that bad, as eventually I do learn. Unless a person has extremely distinctive features (say, three noses), it takes me about a dozen times of meeting them before I can confidently identify them in a small crowd, though larger crowds and unfamiliar settings throw me off.

    If (despite the fact that I'm not as far gone as those extreme cases) you think it'll further science, I'd definitely help out!

  6. I did about the same in the practice as the trial. When I was wrong, it was usually me uncertainly guessing and suspected I was wrong. Anyway, me score was: 3.12

  7. For short term visual recognition, I have to rely on distinguishing feature recognition. As near as I can tell, I code a specific feature into a concept mentally and then I can use that to distinguish. This only works when the pool is fairly small (ie, cast members in a movie), so that I can remember unique characteristics. For long term and more permanent facial recognition, I have no clue how it works; it seems to be a very different process.

    In the case of the practice run, the characters were all about the same to me. Bulbous, kind of like a bird, with small pieces jutting off of them. The pieces were all too small and varied to be quickly recognized and identified. I had no way to code those features into concepts. The actual test characters were easy for me to code into a word/concept as they mostly very closely matched a well known shape. Apparently I am able to say "dog" over and over and still not force those coding words out of my mind.

    I'm not as bad about faces as what Rebecca mentioned. I, too, eventually learn and recognize people, but even then it's pretty contextually dependent. If I'm not expecting to run into someone and I don't know them really well, it's quite likely I won't recognize them at all. Basically almost exactly how Rebecca described it.

    I simply don't seem to have much of a visual memory in general. I remember almost everything in phrases or descriptions. If I picture my mom's face or a house I lived in I usually picture an abstract version of the distinguishing characteristics (for example, a featureless face with long brown hair). I am NOT the person you want witnessing a crime. :)

    vacognition, I'll send you an email offline. I'd be willing to participate in online experiments in this area.

  8. I simply don’t seem to have much of a visual memory in general. I remember almost everything in phrases or descriptions. If I picture my mom’s face or a house I lived in I usually picture an abstract version of the distinguishing characteristics (for example, a featureless face with long brown hair). I am NOT the person you want witnessing a crime.

    My memories are just like that, too! I describe it as an impressionist painting. I suspect that this is one reason why I can't draw very well…I try to picture something in my head, and then try to zoom in on the details to recreate them on paper and I'm just completely unable to do it. Even though I can picture, say, Mickey Mouse in my head, when it comes down to it I can't tell you just what his nose looks like.

  9. What's interesting is that, while I actually am pretty good with faces, I estimate my ability much lower than it is. So often I'll see somebody I know and recognise them but brush it off. And then later when I look back and realise that it is who I thought it was, I feel silly.

    The problem was compounded when I was taking ballroom dance classes, because of course girls always outnumbered guys 3:1 at least, and through the mechanics of partner rotation I'd end up dancing with pretty much every girl in the class. (Not that I had any complaints!) This would result in some awkward situations later when I'd bump into a girl who clearly knew me and who I vaguely recognised but couldn't remember anything about.

    Also, even worse, I tended to recognise people I've never been introduced to but merely see around sometimes, either from the dorm or the T or wherever. I never quite knew how to deal with that, because my instinct on seeing someone I recognise is to say hi, but social conventions being what they are that could be weird, especially if the other person doesn't share the same capacity for recognising random strangers.

    So I guess the moral of the story is that social interaction is weird.

  10. Oh, I guess I should add that there's a rather obvious bias in my face-recognition. I'm far more likely to remember female faces than male ones.

    In fact, I've noticed that I'm actually prone to false-positives in a lot of cases, e.g. girls I have a crush on. Again, the reasons for this should be obvious. ;)

    I suspect that the phenomenon should be fairly general, although I wonder if there's any data out there about this kind of bias in face recognition.

  11. Yes, that faces out of context thing can really throw you. You can see your baker for years behind his counter and chat with him, but not recognize him when you meet him in the park.

    And then, even if you recognize someone, you still have to figure out the right name to go with the face…

    also did terrible on the practice, where I tried to look at each shape on it's own. With the black shapes I looked at the whole thing at once to see it as one big symbol, and then I could 'feel' if one bit was out of place. Still only got me a 3.111, mind you…

  12. I was actually relieved when the actual test started because I found the shapes to be much easier to distinguish. With the practice run, I was wrong almost every time, and the rest were guesses.

  13. Only a 2.95156695156695 for me. I found my dogging to lag a bit as well, but it also just faded into the background and I still was able to apply names to the shapes in the actual test at the same time as I was saying 'dog'. Not sure if that throws my results off, but there you go.

    I don't have a particular problem with faces, and actually I'm quite good at knowing people (identifying actors in films is something I'm usually very capable of doing, etc). However, I tend to distrust my face recognition sometimes, and will be quite hesitant to talk to people if I've met them only a couple of times and might still be 'wrong' about who they are.

    I'm even worse, however, with names. I am not usually incorrect with names, but I NEVER use them in conversation because for some strange reason I have it in my head that I've got the name wrong. I do this even with some very good friends. Not much to do with visual memory, I realize, but odd nonetheless.

  14. Damn, everyone did better than me. I only got a 1.97 and some more numbers. But I did manage to say "dog" all the way through, and was extremely conscious of doing so, to the point where I found it distracting. Which probably had something to do with it.

    I'm not bad at remembering faces, but I'm awful at names, which always leads to awkward conversations, as I franticly phrase everything so as to not have to say the person's name. Still, I've gotten pretty good at that.

  15. I got a 2.45 or so. I remembered to repeat "dog" the whole time, as well. I had no trouble with the "dog" part, mostly I think because I stuck with a simple rhythm and phrasing pattern, which does make it easier to perform a repetitious task — for me, anyway.

  16. I got a 2.2something-or-other… which is okay, but my roommie started taking pictures and the camera flashing was an unwelcome distraction. I looked away a couple times only to have to guess… wrong. I had no trouble saying dog over and over, because it kind of rolls off of my tongue when repeated (and I often talk to myself, so it's nothing unusual or distracting)… interesting test.


  17. 3.8 for me – woohoo – and I was very pleased with myself until I realised I had forgotten to say "dog" to myself. My labeling of the different pictures probably helped a lot with the high score.

    Like some other people have mentioned I am good with faces but horrible with names so it doesn't do me much good unfortunately. "Hi! Nice to see you again – we met in the gym – what was your name?"

  18. I got pi. (3.14…)

    My "dog" mantra lagged a few times, though. I think the test was tricky enough without that extra handicap.

    And I'm pretty good with faces and names, though, like Joshua, I'm a little better with the females.

  19. I read this article awhile back on prosopagnosia (face blindness). I'm actually very good at recognizing names and faces it's a wonder I didn't go into sales instead of software development. My wife on the other hand might remember your face but she'll never remember your name. She calls me Rainman when I will recognize people *she* used to work with.,9171,…

    By the way, 3.6666666667

  20. 2.8 and change. I think I might have done better if I wasn't in such a hurry to answer each one though. I sort of got into a rythm trying to match the pace of the automatic changes and "knew" several times I had the wrong choice before clicking.

    And I said dog-dog-dog-dog-dog-dog-dog-dog continously the whole time.

  21. Well, looks like I'm the worst… I got 0.4 seconds, I think I'll try it again and cheat by giving stuff a name so I can feel good about myself; in general I have a very fine memory, oh well time for some chocolate and coke I think.

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Back to top button
%d bloggers like this: