Are you tone deaf?
February 5, 2019 11:02 AM   Subscribe

Test your tone-deafness with The Music Lab at Harvard University's department of psychology.

More games/tests from the Music Lab:
  • World Music Quiz: Can you tell what a song is used for even when you have no experience of the culture that made it?
  • Who's Listening? Listen to voices from around the world and guess who's listening.
  • Synthesizer Game: Help us to synthesize world music and learn why songs sound the way they do.
via

Also from the Music Lab: Can You Tell a Lullaby From a Love Song? Find Out Now
posted by not_the_water (95 comments total) 39 users marked this as a favorite
 
I wish they had a 'can't tell' option, because on the tonedeafness quiz I probably was guessing about a third of the time.
posted by dinty_moore at 11:14 AM on February 5 [9 favorites]


Ditto on wanting a 'no clue' button. Got 25 out of 32 anyway.
posted by YAMWAK at 11:16 AM on February 5 [2 favorites]


I did average! Yay me!
posted by aubilenon at 11:16 AM on February 5


I did pretty well, but ha - that calibration for hard-of-hearing people sounded like a recording from The Gates Of Hell.
posted by queensissy at 11:17 AM on February 5 [11 favorites]


I'm basically tone deaf! I did better than 16% of people...by guessing!

This is gratifying, because people have always made fun of my singing voice, I had a horrendous five years of music lessons which taught me nothing, etc etc, and this has always been attributed to some willed failure on my part...but it turns out that I am just intrinsically amusical, which is what I'd suspected all along.

This leads me to wonder what I'm enjoying when I'm enjoying music, because it obviously isn't the tones.
posted by Frowner at 11:19 AM on February 5 [5 favorites]


You listened to 32 sounds. Of those you guessed 18 correctly!
Nice work! You did better than 2% of people
Good job me.
posted by jeather at 11:19 AM on February 5 [7 favorites]


Nice work! You did better than 11% of people.

The "Nice Work!" makes it sound even worse than it was.
posted by Bulgaroktonos at 11:19 AM on February 5 [8 favorites]


I did quite well (31/32 - you wouldn't know it to hear me sing), but I really wanted to know how close the tones were for the one I got wrong.
posted by Jeanne at 11:20 AM on February 5 [6 favorites]


30/32. Yeah, there should be a "didn't change" option.
posted by Halloween Jack at 11:22 AM on February 5 [4 favorites]


Yeah I got 28 and I really want to know the frequency difference on those I missed and the 2 or 3 that were lucky guesses. I think I'll try again with my much better headphones at home.

I was hoping for a "name the note" kind of thing but this was fun.
posted by Clinging to the Wreckage at 11:24 AM on February 5


I'm hard of hearing and I got 27 right (man, some notes were close) so tone-deafness is not one of my hearing issues. But yeah, the calibration test was kind of terrible. Mostly just sounded like directional changes in sound (coming from the front or the rear) making it difficult to determine if it was louder or softer.
posted by acidnova at 11:25 AM on February 5


You listened to 32 sounds. Of those you guessed 19 correctly!
Nice work! You did better than 3% of people. Your average speed was 1.5 seconds.


Yay?

I wish they had a 'can't tell' option

Agreed. I guessed on a lot of them.

I'd like to know if I was better at detecting lower or higher notes.

I want to print this out and hand it to people who tell me I just need to practice more before I can figure melodies out by ear. I've been trying for 30 years, I think I've practiced enough.
posted by bondcliff at 11:27 AM on February 5 [2 favorites]


I got 31, which makes up for my unmentionable score on the UI test this morning.
posted by theodolite at 11:29 AM on February 5 [1 favorite]


A few years ago I organized a library reading by Tim Falconer, the scientifically tone deaf author of Bad Singer, and it was fascinating to hear him speak about how he processes music and how it differs from the average person (the book was interesting, too). A friend of ours is tone deaf as well, and the first time I heard her sing I thought she was trolling everyone by singing off-key on purpose. I'm not much of a singer at all, but I'm not sure I could miss that many notes on purpose while singing along to prerecorded music because the natural tendency is to try and correct as you go along.
posted by The Card Cheat at 11:31 AM on February 5 [4 favorites]


I am deaf to some tones, just not those ones.
posted by drlith at 11:36 AM on February 5


29, but one of the ones I got wrong I just glitched and pressed the wrong button, while the other two sounded exactly the same as the reference tone to me. It said I did better than 59% of people.

I knew I wasn't tone deaf, but I am absolutely abysmal at singing.
posted by knapah at 11:41 AM on February 5 [2 favorites]


If you want to do well and are not in a quiet place, turn up the volume until it's a lot louder than you'd think you'd want it to be based on that reference noise.
posted by sfenders at 11:47 AM on February 5 [2 favorites]


98%
posted by polymodus at 11:47 AM on February 5


I think my violin lessons up to middle school, at that age, where we played in orchestra and learned to tune our instruments, made a difference especially with the small changes in tune.
posted by polymodus at 11:50 AM on February 5 [3 favorites]


Nope, not tone deaf (34/34, also have absolute pitch).
posted by Dashy at 11:50 AM on February 5 [4 favorites]


The 'can't tell' ones were different, but were below your pitch perception ability, which was rather the point of the test. I suspect they can tell which ones you're unsure or guessing on because your response time usually goes up for those - it was three times as long for me.

29/32 on speakers with only a couple of lucky guesses, which was a bit of a relief considering I know my upper hearing range is disappearing rapidly as middle age sets in; so I was wondering how badly I was losing the ability to hear different tones in the middle ranges. Far from absolute pitch myself, but as an ex-musician a big part of it is learning relative pitch, so disharmonious singing still sets my teeth on edge.
posted by Absolutely No You-Know-What at 11:52 AM on February 5 [2 favorites]


I love tests like this. They're testing various things. Your basic ability to discriminate tones. Your reaction time to discriminate tones, probably correlated to the distance between them. (At least for me; half the samples I could identify in 0.4s, the other half were 1s+.) And then possibly also a priming effect; if the last 3 tests were all D, are you more likely to hear D the fourth time?

Agreed there's probably no "these were identical" samples, just some samples that are too subtle to detect. They'll see a 50% error rate on those if truly no one can tell them apart.
posted by Nelson at 11:54 AM on February 5


Nice work! You did better than 16% of people

Thanks for letting me down easy, but my math is better than my musical ability: this means I did worse than 84% of people!

I feel like I might have improved during the test. Maybe I will try it again with more volume.
posted by exogenous at 11:54 AM on February 5 [2 favorites]


The test was centered at A440, and no differences were larger than a half-step, which would be ~25 Hz.

I don't know what the minimal frequency difference was, but my ears say 1 Hz or less.
posted by Dashy at 11:54 AM on February 5 [5 favorites]


Always thought I was totally tone deaf but 28
posted by Cosine at 11:54 AM on February 5 [2 favorites]


31/32.

The odd thing is that the one I got wrong I simply couldn't tell, but -all- of the ones I couldn't tell after that point I said "up" and was correct, which means I am much better at telling if a note is flat than if it is sharp - something I didn't know.
posted by solarion at 11:55 AM on February 5 [6 favorites]


I can sing very well. And I still got 23/32. All of the ones I got wrong I thought were too close to tell, so I guessed.


And I guessed wrongly!
posted by droplet at 11:55 AM on February 5


It's 2019, this is a Harvard thing, and you can't do it on an iPad.
posted by sageleaf at 11:55 AM on February 5 [4 favorites]


The test was centered at A440, and no differences were larger than a half-step, which would be ~25 Hz.
I'm pretty sure I heard some larger than that

***
It really amazes me how many people seem to thing that singing ability is equivalent to ability to discriminate pitch. I have a reasonably good relative pitch sense, and a singing range of about three half steps.
posted by thelonius at 11:57 AM on February 5 [4 favorites]


You listened to 32 sounds. Of those you guessed 23 correctly!
Nice work! You did better than 11% of people. Your average speed was 0.8 seconds.


Not quite as tone deaf as I thought. I only sing when there is absolutely, positively, no one else around. Unless I've been drinking.
posted by antiwiggle at 12:08 PM on February 5 [1 favorite]


I apparently assumed they would all be half or whole steps away. The first couple of very close tones threw me. I panicked thinking "Where's the 'same' button??" until i realized that they were just a little bit off. 29/32.
posted by scottatdrake at 12:10 PM on February 5 [3 favorites]


I missed a couple of them because I was trying to go too quickly and hit the wrong button, and still beat out 59% of people. Go figure.

I was also waiting to have it go 3 - 2 - 1 and then have some horror movie monster pop onto the screen with a hideous scream and inform me that I'd been pranked.
posted by delfin at 12:10 PM on February 5 [2 favorites]


Is this an experiment where you're told one thing's happening but in fact something else entirely's being measured? For example, if it's a tone-deafness test, why are the times of one's reactions to U/D noted/important at all? Enquiring minds want to know!
posted by riverlife at 12:12 PM on February 5 [1 favorite]


31 out of 32. I felt the one I missed didn't change.
posted by bwvol at 12:18 PM on February 5


if it's a tone-deafness test, why are the times of one's reactions to U/D noted/important at all?

My guess is that if you usually respond in, say, 1 second, but there were a few you only got right (or wrong) after hesitating 2-3 seconds while you decided, that factors into their statistics somehow.
posted by Greg_Ace at 12:20 PM on February 5 [2 favorites]


Where did you all get your results? I didn't see mine, must've missed it, darn.
posted by iamkimiam at 12:22 PM on February 5


Is this an experiment where you're told one thing's happening but in fact something else entirely's being measured? For example, if it's a tone-deafness test, why are the times of one's reactions to U/D noted/important at all? Enquiring minds want to know!

I suspect it's at least partially way to get you to go with your first perception/instinct, rather than analyzing deeply and mentally replaying the tones (or whatever) until you're sure. Make people feel rushed and they'll make fast decisions.
posted by Hold your seahorses at 12:22 PM on February 5 [1 favorite]


I apparently assumed they would all be half or whole steps away. The first couple of very close tones threw me. I panicked thinking "Where's the 'same' button??" until i realized that they were just a little bit off. 29/32.

Same score, same problem. I have fairly good "tuning" pitch (i.e. "is this slightly off the tone one way or the other") from years of playing instruments, but that's not what I thought I was listening for, and by the time I figured out that the "same" tones weren't a test malfunction but rather the whole damn point, it was too late to redeem myself.
posted by Hold your seahorses at 12:25 PM on February 5 [1 favorite]


28 out of 32! Huh. I've always equated musicality with tone deafness, which I now understand is completely incorrect. I don't sing around other living creatures because I realize that I sound horrible -- I guess the insult is that tone deaf people sing badly because they don't realize they sound bad? Hmm. It's actually an insult based on some generous assumptions.
posted by grandiloquiet at 12:27 PM on February 5 [1 favorite]


"For example, if it's a tone-deafness test, why are the times of one's reactions to U/D noted/important at all? "

It's so they can correlate reaction times with correct responses AND people's own experiences and their perceptions of their tone deafness. A hypothesis would be something like, 'We expect people who don't think they are tone deaf and have had experience with music/being sung to as a child, etc. to respond quicker and more accurately.'
posted by iamkimiam at 12:31 PM on February 5 [1 favorite]


I found that when I hesitated and replayed the sounds in my head, I got more right than wrong. Weird.

(22)
posted by joeyh at 12:34 PM on February 5


I am definitely not tone-deaf and I noped out after getting so focused on the timing aspect that I tried to guess what the next note would be in order to get a sub-0.3s reaction time and guessed incorrectly.
posted by grumpybear69 at 12:34 PM on February 5 [2 favorites]


30 out of 32 - oddly, the ones that were slightly flat were easy to detect and the ones that were slightly sharp weren't - they sounded the same to me
posted by pyramid termite at 12:40 PM on February 5


I scored in the 80th percentile, but also can't sing very well. When I sing it sounds pretty good in my ears, but listening to recordings indicates that I'm fairly flat. I've considered taking voice lessons before, because I really do love singing and would like to be better at it, but I've also considered the fact I might be tone deaf. Is there anything that would account for that discrepancy in terms of biology, or is that something that could be corrected through training?
posted by codacorolla at 12:43 PM on February 5


It really amazes me how many people seem to thing that singing ability is equivalent to ability to discriminate pitch. I have a reasonably good relative pitch sense, and a singing range of about three half steps.

I mean, me too, but the idea that decent pitch discrimination is required to sing well (i.e. to sing on pitch) makes some intuitive sense, whereas nobody would assume that having a pleasing voice is required to develop pitch discrimination.
posted by atoxyl at 12:45 PM on February 5


29/32, and all three I missed were in the front half; after that I closed my eyes and was a lot better at catching the close ones (though since it's randomized, I guess it's also possible I just didn't get any in the second half quite as close as those three I missed).
posted by solotoro at 12:49 PM on February 5


It really amazes me how many people seem to thing that singing ability is equivalent to ability to discriminate pitch. I have a reasonably good relative pitch sense, and a singing range of about three half steps.

Okay, but if I can't hear tones, doesn't that play a part in being unable to sing in tune? I can't self-correct, because I can't hear the difference between what I'm hearing and what I'm singing.

Believe me, the problem with my singing voice is not range.
posted by Frowner at 12:53 PM on February 5


Am I tone deaf? Actual deaf, thanks!
posted by Capt. Renault at 12:54 PM on February 5 [6 favorites]


I mean, me too, but the idea that decent pitch discrimination is required to sing well (i.e. to sing on pitch) makes some intuitive sense, whereas nobody would assume that having a pleasing voice is required to develop pitch discrimination.

Sure! But people seem to think that, if you can't sing, that's being "tone-deaf"
posted by thelonius at 12:55 PM on February 5 [3 favorites]


oddly, the ones that were slightly flat were easy to detect and the ones that were slightly sharp weren't - they sounded the same to me

That seems to be a fairly common thing. I've always been able to tell when someone (or myself) is playing flat; but if it sounds out of tune yet clearly isn't flat, then I have to assume it's sharp.
posted by Greg_Ace at 12:56 PM on February 5 [1 favorite]


Am I tone deaf? Actual deaf, thanks!

I like very much that Ameslan was one of the options for native language.

27/32 for me, but at least eight were pure guesses. Okay for years of factories and playing in bar bands, I suppose.
posted by ricochet biscuit at 1:05 PM on February 5


Nope, not tone deaf (34/34, also have absolute pitch).

Me too! :) Secret handshake!
posted by Melismata at 1:09 PM on February 5


Only missed one - I guess I can't blame my out of tune singing on being tone deaf apparently. Though when I'm trying to match tones with someone, I'm awful at picking out whether my own voice is sharp or flat.
posted by Zalzidrax at 1:13 PM on February 5 [1 favorite]


30/32. No training beyond childhood/teenagehood music lessons (clarinet and piano, mediocre in both, in part because I haaaaaated practicing).

It's interesting; I can carry a tune, I guess, though my voice quality is meh at best. But listening to other singalong randos who think they're in tune can be... painful. I mean, you're having fun, so I'm not mad or anything - but I can still hear it.

So a pretty good result doesn't surprise me.

That said, trying to stay in tune always feels more like a matter of controlling my vocal cords than a matter of what I hear. And guesswork / pattern matching. I still play Rock Band sometimes with vocals (so, scored karaoke), and when I manage to pass a song I've never heard before, my spouse seems to think this is some kind of magic trick. I'm like, IDK, I hear what key it's in (even if I can't name it) and so I can guess where the uppydowny line is going? Plus you hear a verse and a chorus and you're set until the bridge, at least in a typically structured pop song.
posted by cage and aquarium at 1:13 PM on February 5


I tried this with a one-sided headset so you don't have to. Gah.
posted by wellred at 1:14 PM on February 5 [1 favorite]


93.56 percentile. Woo.

I missed three; I couldn't hear any difference between the two tones in those. The first time I couldn't hear any difference, I guessed up, but it was actually down, so the next two times I couldn't hear any difference, I guessed down. I was right both times. Which is actually sort of interesting, because I do tend to sing sharp, so maybe I just tend to hear a bit higher than the actual tone.
posted by holborne at 1:17 PM on February 5


32 out of 32. It said I did better than 100% of the people. I guessed the one that sounded the same. Semi-tone difference? Not likely. Average time one second. Now if I could only apply my skill to singing.
posted by njohnson23 at 1:21 PM on February 5


29/32 with only a few guesses. Halfway through the test it seemed to get easier; I'm not sure if this is by design, statistical fluke, or me learning on the fly to better distinguish between minute variations in pitch.
posted by chrominance at 1:22 PM on February 5


I'm pretty sure I'm not that tone deaf, but I'm apparently more tone-deaf than a lot of the other people who have taken this test.
posted by zennie at 1:25 PM on February 5


32/32, though the first "close one" had me looking for the "same" button, and I ended up guessing on that one. I think the interesting finding, for me, was that after practicing finding the way-sharp or way-flat tone a bunch of times, I was suddenly deaf to a small change -- not even noticing it.

It really amazes me how many people seem to thing that singing ability is equivalent to ability to discriminate pitch.

Mostly it means that I know I sing really out of tune.
posted by klausman at 1:33 PM on February 5


30/32. I never really did a lot with music except of one miserable year of 7th grade band, but I grew up in an a cappella Christian tradition and boy, were we ever serious about hitting the right notes. That probably helps. That's basically three hours a week of choir practice beginning at age zero.
posted by Pater Aletheias at 1:42 PM on February 5


First time around, I was listening for differences in the three "reference" tones (they're modulate a little) and completely ignoring the last note which is...rather different in most cases.

Once I turned it up a little (I didn't follow the directions to make it loud but not uncomfortable; I just left it medium loud), I got 31/32, and the one was a mis-type.

Normal hearing in one ear, -30dB + hard cutoff above 1kHz in other ear. My bad ear is very sensitive to out-of-tune stuff below the cutoff, though.
posted by notsnot at 1:42 PM on February 5


All of this makes me wonder what does make a good vocalist? Because I did "OK" in this test and the less said about my singing, the better. I mean, among my singing flaws:

* I have no range, to speak of
* I cannot consistently hit a note
* I do not have any power
* I cannot hold a note for any length of time
* I don't seem to have my own "voice", I cannot sing a song without adopting the "style" of a version I have actually heard (ie, if I sing a Johnny Cash song, I end up trying to sound like him, etc.)
* To sing even passably in the shower I have to consciously adopt some sort of "accent", like a pseudo british, southern us, or australian accent

I mean, that's just a hot mess, but I would really like to know what the last one is about.
posted by maxwelton at 1:42 PM on February 5 [4 favorites]


Mostly it means that I know I sing really out of tune.

I wonder if there is a curve that does the opposite of what we might expect - that for many to most people as pitch discrimination (relative, absolute, just a 'sense' of it) improves their likelihood of singing in public decreases. I sing like mad in my car and I've had people tell me that I can sing well but (since I quit drinking) there is no chance I'm singing in front of people. I can tell when I'm off even a little bit and it's grating. I don't want to subject others to that.
posted by Clinging to the Wreckage at 1:45 PM on February 5


I got a 30/32, which was better than I thought I'd do. I started my office's vacuum robot right before jumping in which probably didn't help.
posted by geegollygosh at 1:46 PM on February 5


It said I did better than 85% of people. That kind of surprises me as I absolutely cannot sing in tune, and have never been able to grok playing music on any instrument. Admittedly, there were a few in the test where I simply had to guess.
posted by Thorzdad at 1:59 PM on February 5 [1 favorite]


32/32. Here I thought it was going to be like guessing intervals and it was mostly sneaky fraction tones. Still surprised I got them all right though.

( I'll confess that I guessed on the one I was pretty sure didn't change, but there was no option for same/same.)
posted by Lizard at 2:07 PM on February 5


It really amazes me how many people seem to thing that singing ability is equivalent to ability to discriminate pitch.

It is, in fact, quite the opposite. In music school I was in a sight-singing class and it became apparent very quickly that instrumentalists generally had good pitch but middling-to-poor vocal tone while voice majors had great vocal tone but generally middling pitch.
posted by grumpybear69 at 2:11 PM on February 5 [2 favorites]


I'm a little baffled about all the comments about singing here. Singing in tune, even when you can hear the difference is a physical challenge. You can have a great ear but still poor technique when it comes to making your very complex systems of vocal folds actually do what you want them to. As anyone who has ever tried to sing when they felt even remotely sick can attest to. And hearing external sounds is a different experience from hearing sounds produced by your own body due to your skull acting as a resonating chamber (and for those less experienced/trained) can cause some confusion in your perceptions.

Another point - we are only being asked to say whether a given note in a sequence is higher or lower than the note immediately preceding. In addition to being hard of hearing, I also have poor eyesight but when ever I do those color blindness tests, I can easily discern that two colors are slightly different shades from each other. But depending on the lighting in a room, struggle to be able to differentiate more radically different colors sometimes. Such as, "get the blue shirt" and I grab the green one because they weren't right next to each other for me to see that "oh, that one is blue-er." And I'm not talking weird lights but just lights that are dimmer than is optimal for me.
posted by acidnova at 2:12 PM on February 5 [3 favorites]


Or a long-winded way for me to say that lab settings and real world settings can produce very different results. This test is just trying to isolate a very specific skill. How well can one determine a very minute change in frequency? And these are pure wave tones as well.
posted by acidnova at 2:15 PM on February 5


Generally, I have good hearing, but I'm worse than average at discriminating a voice in a noisy setting. Because of this my parents had my hearing checked several times, which amounted to sitting at a desk with headphones on and raising my right or left hand whenever a tone played in the corresponding ear. I always passed that test with flying colors, since it had basically nothing to do with the situation that caused me trouble.

Anyway, I got 31/32 which is probably little comfort to the high school band teacher who placed me as second to last clarinet.
posted by ckape at 2:40 PM on February 5 [1 favorite]


The first time, my score was 24%. I took the test again, only this time I turned the volume WAY up, and improved my score to 85%. Volume really matters for detecting small changes.
posted by Transl3y at 2:47 PM on February 5 [1 favorite]


32/32 and I have some hearing impairment (tinnitus and I definitely don't hear as well as I used to).
posted by cooker girl at 3:29 PM on February 5


A realization I came to a few years ago was that there are people who can hear music in their head. They have a mental stereo they can turn on and listen to at will. They can sing songs just by making the music play inside them. They don't need to memorize lyrics as the words are automatically part of the music and embedded there. I have had the experience of music clicking on in my head a couple times in life and each time it was really amazing. But it isn't something I can control or muster. It made me realize how a deaf Beethoven could compose music. He could hear it all in his head.

I played piano and cello and gained a technical proficiency with the instruments (read, count and place hands and bow to play what is written) but can't make music because I don't hear music internally. There was one guy in my orchestra who one day began playing, Michael Jackson's "Beat It" on his cello and I was stunned. Wondered how that was even possible as the song just came out and no one had sheet music for it. When I asked him how he learned it, he basically shrugged and just said he was playing what he heard. I didn't understand until many years later when I first experienced hearing music in my head. I asked one of my friends who is fairly musical about it and he thought that being able to play music in your head was simply some normal ability that everyone had.

On the other hand I can see pictures in my mind with as much detail as real life or a movie. When I have a blank sheet of paper I can already see what I want to draw and just copy it from my mind. I can conjure whatever I like and manipulate colours, forms and figures at will. I thought this was normal and an ability everyone had until I read an article posted to MetaFilter about different levels of mental visualization and realized there is a wide variance in how we all see / hear and process the world.

So auditory actueness shouldn't really correlate to music / singing ability. I really feel that you can train and be technically proficient and hearing does help on that end but without an internal jukebox you will always be limited.

I did well on the test but felt I just did a lot of good guessing.
posted by phoque at 3:42 PM on February 5 [11 favorites]


I remembered that mefi post so I found it for anyone reading phoque's comment who might be curious to know what's being referenced. It was so fascinating to learn about the very wide and diverse spectrum of abilities to mentally see and hear.
posted by acidnova at 3:58 PM on February 5 [1 favorite]


32 out of 32. It said I did better than 100% of the people. I guessed the one that sounded the same. Semi-tone difference? Not likely. Average time one second.

My 27/32 above was indifferent but my average response time was 0.5 seconds. Years as a white collar worker in the private sector has taught me the importance of getting the wrong answer quickly.
posted by ricochet biscuit at 4:17 PM on February 5 [2 favorites]


Semi-tone difference? Not likely.

I can definitely hear quarter-tones (which are 50 cents apart), so the really close ones had to have been within only a few cents of each other.
posted by Greg_Ace at 4:29 PM on February 5


I'm a pro musician and got 30/32. I'm so ashamed.
posted by fast ein Maedchen at 4:49 PM on February 5


its funny, Singing is very intuitive, and much of the sound science involved seems almost counter-intuitive at times, which is interesting. Context is important, which is why some singers with imperfect intonation succeed at conveying emotional feelings. (29).
posted by ovvl at 4:52 PM on February 5


..."there are people who can hear music in their head. They have a mental stereo they can turn on and listen to at will."

This is (sorta) me. I did a vaguely related ask.me some years ago and was gifted the phrase "cranium radio." Sometimes I really, really wish I could turn the dang thing off, as the music I "hear" is not often under my control.
posted by goofyfoot at 4:55 PM on February 5


The percentile score they give you weighs your response time heavily. I got 29 correct the first time but, because I was quick on all but the most confusing one (the first small increment after a bunch of large increments), I was told that I did better than 84% of people. Second time I got all of them correct and was told I was better than 100% of people, when one would expect that being better than 99% of people would be the maximum possible.

I do not have perfect pitch, I have not played an instrument for more than a few days in over four decades, and until a recent stereo purchase almost never listened to music. But when I was a kid and used to go to church, I was able to sing by sight-reading music. Not a great singer, but I can hear the notes.
posted by chromium at 6:09 PM on February 5


I got 27/32. Like a lot of us, I thought the ones I missed and some that I didn't miss sounded damn identical to the reference tone, and this made me curious how close they were. Since the Music Lab experiment doesn't tell you that, I wrote my own program to emulate the experiment as closely as I could (three half-second 440 Hz reference tones followed by a higher or lower tone), but I wrote mine to tell me after each round how many cents difference there had been. After a bit of testing I realized that differences larger than 20 cents were never going to trip me up, so I narrowed the range to produce differences randomly distributed between –20 and +20 cents, then I played 100 rounds. (Though at one point I had to start over because some Lunar New Year firecrackers started going off in the street!)

My results: 96/100. The 4 times I missed, the actual differences were –0.6, –1.4, –3.5, and –1.5 cents -- all flats that I misjudged as sharp. But I actually got 11/14 even when the differences were less than 3 cents, so I'm going to say that some of those Music Lab tones were, if not damn identical, at least really damn close to the reference tone.

What interests me is that, when the differences were very small (under 5 cents or so), my level of confidence in what I had heard seemed unrelated to the exact size of the difference. Sometimes I felt like I was guessing, then the difference turned out to have been 4 cents; other times I felt perfectly sure of myself and the difference turned out to have been 1 cent. This observation renews my curiosity about what the Music Lab study is learning from the response times.

By the way, I am happy to share my program if there is any demand, but it's in Chipmunk Basic which might limit its appeal. :-)
posted by aws17576 at 6:43 PM on February 5 [4 favorites]


Jukebox mind here. I have long wished I could just record my mind music directly to tape because it would be so much easier to sketch out songs, though I suspect if I could do thay I would realize how incomplete my mind arrangements really are.

The most memorable mind music moment I ever experienced was maybe 16 or 17 years ago when I was lying in bed and imagined a shredding guitar solo. I could make out every single 32nd note. It was amazing and felt virtuosic.

Images, on the other hand, are sometimes vivid and sometimes small and grainy, but always keep morphing and changing no matter how hard I try to keep them stable.

I heart brains.
posted by grumpybear69 at 6:48 PM on February 5


Years ago I had a vivid detailed dream of an absolutely gorgeous arrangement of a Beatles tune for a stage band (woodwinds, trombones, trumpets, and a rhythm section). Sadly, as usual, by the time I was awake and reasonably coherent almost all of it had permanently vanished from my memory.
posted by Greg_Ace at 7:11 PM on February 5 [1 favorite]


30/32, and then 32/32 with my eyes closed. If you want to try again, turn it up loud, put one finger on "d" and one finger on "u" and close your eyes before you start.
posted by tzikeh at 9:47 PM on February 5


noise cancelling headphones seem to have an effect. The first time I did the test I got 27/32, (but there were about 8 where I felt was guessing). I did it again with noise cancellation turned off and got 32/32 (and was completely certain about all but one)
posted by lastobelus at 11:03 PM on February 5


I don't know what the minimal frequency difference was, but my ears say 1 Hz or less.

+/- 1 Hz from 440 is a little under 4 cents. I've seen it stated that the minimum audible pitch difference is around 5 cents. So yeah, it's probably close to that.
posted by atoxyl at 12:37 AM on February 6


first time, I got 29; the second time, a little louder, 30. One, that time, was a fat finger. But with the second time I chose "up" whenever I could not hear that the tone was definitely lower, and this worked.

What's weird is that I have a kind of synaesthesia about tones. Lower are always darker, and higher ones brighter: chords suggest complex *shapes* to my mind, and particular instrument tones have their own colours and sometimes shapes -- the first time this struck me I was listening to the Beatles' "Badge" aged 14 or 15 and there is a particular guitar phrase in the solo which has the clearest imaginable bronze flecked with gold character to me.

This processing is both vivid and inaccurate: when I was doing the test I had to use a kind of blindsight to tell the tones apart because the way I "saw" them was very often wrong. I would hear objectively lower tones as brighter and to get the right answer I just had to disregard this.
posted by alloneword at 1:56 AM on February 6 [1 favorite]


Funny, I had the opposite experience: higher tones were very obviously higher, while lower tones gave me pause.
posted by alexei at 2:08 AM on February 6 [1 favorite]


I was a little surprised to find that I couldn't give instant answers to the microtonal ones but was able to get most of them after replaying the sequence in my mind. I guess I know what's going on better when I have to produce it instead of just hearing it, even if I'm not really producing it. I am good at hearing music in my head so it is fascinating to read the comment above about what it's like not to be able to do that.

I remember being mortified as a kid taking essentially this same test as part of a choir audition and giving the wrong answer to one of them. It was using a piano instead of a tone generator, so I didn't have the excuse of unnaturally close pitches.

27/32 (45th percentile) with speakers, 30/32 (85th percentile) with headphones.
posted by enf at 7:59 AM on February 6


32/32. I found the really close flat ones easier to tell than the sharp ones. It'd be nice to get a fully detailed report in the results. I definitely don't have perfect pitch though, I can't even imagine it, is it like how a particular wavelength of light is just obviously that colour?

Cool to see Chinese differentiated out a bit in the language options, I guess they're looking at the tonal language correlation.
posted by lucidium at 3:40 PM on February 6


There is absolutely no point for me to take this without a "no difference" button. But then I suppose there's nothing for me to gain beside an abysmal score anyway.
posted by dances with hamsters at 6:10 PM on February 6


The omission of a “can’t tell” option is deliberate.

The tone-deafness test is a psychophysical experiment. “Forced choice” is absolutely standard in these; the operating principle of these experiments is not “tell me what you can’t do”, but “prove to me what you can do”. The aim here is to determine just-noticeable differences, and for this purpose you want your subjects to randomly guess if the task is indeed impossible for them to solve; it’s much more reliable than letting them consciously select whether they can or cannot tell the difference. The experimenters will be able to see, based on your error rates, that you cannot do the task when the difference between the reference stimulus (the three beeps) and the test stimulus is smaller than your individual threshold. They might, for example, fit psychometric curves to the ratios of correct responses as a function of pitch difference: For each subject, they will see ~100% correct answers for the largest differences and ~50% (random guessing) for the smallest, and they’ll be able to work out where your actual threshold lies based on that. Offering a third option would only make their data more complicated to analyse.
posted by wachhundfisch at 4:49 AM on February 7 [8 favorites]


Yeah, on the aggregate, it's probably fairly useful for them, but for the purposes of me wanting to know my personal results, it's still frustrating - I don't trust my score because I'm aware of how much noise there is in my personal results. Since this test does promise to let you know your personal results at the end (and that's the reason why I'd spend my time on it), it's disappointing.
posted by dinty_moore at 6:12 AM on February 7


I'm interested to hear that many of you think you respond more slowly when it's hard to tell them apart.
Me, that's when I respond fastest. I've got no hope of telling them apart, so whatever, press a key, let's move on.

There may be a reason I did worse than 55% of my fellow testmakers. (Hrmph, I can subtract, see?)

That's actually a lot better than I expected, given that a friend of mine once asked why I was trying to sing in apparently at least three different keys...

And also given the computer game Myst. Any of you who remember it well will remember it had a musical puzzle, where you had to play a sequence of tones on a machine in the game. The tones were audible somewhere else in the game, so you just had to remember what it sounded like and then play the tones.
I finally got past that puzzle by diligent trial and error, because I just could not remember the tune.
posted by nat at 11:27 PM on February 12


« Older above it all   |   "You know he’s evil, but the camera sure doesn’t." Newer »


This thread has been archived and is closed to new comments