Comments

1. #1 "Q" the Enchanter

   November 13, 2006

   The test seems much more a test of phrase memory than pitch perception. I’ve got very good relative pitch, as well as what I call a slow-acting perfect pitch (I can identify and generate a given pitch, but I usually need a few seconds to isolate the tone in my mind), but still didn’t score even near 100.)

2. #2 Dave Munger

   November 13, 2006

   Good point, Q.

   Mandel acknowledges this in his description of the task, but claims that it’s a valid test of tone-deafness since most tone-deaf individuals have normal phrase memory. The threshold for a pitch perception deficit is quite low: 50 percent, so if you scored even 60 percent or above, according to Mandel, you probably have normal pitch perception.

3. #3 Dom

   November 13, 2006

   I’m just happy I’m above the median. As a musician, I’ve done quite a bit of aural transcription, under pressure, in formal settings. Picking out the intervals versus stepwise movement, noticing key, etc. On this test, I felt I did better when I didn’t try to think about it, although it was hard to resist for the more familiar ones (the major triad(s?), for instance).

4. #4 etbnc

   November 13, 2006

   Due to some Flash software bug, my test run got stuck in a loop on item 35. That riff is now stuck in my head as an ear worm.

   So, um, thanks…

5. #5 Mondo

   November 13, 2006

   83.3%
   Of course I thought I got them all correct.

6. #6 Giancarlo

   November 13, 2006

   86.1%
   Actually better than I expected (was scared by the comment about excelent musicians barelly scoring over 80%!)
   Sooo… how do I apply to Julliard?

7. #7 Dave Munger

   November 13, 2006

   I emailed the test to a musicologist friend of mine. From his perspective, it’s a memory test, pure and simple, and biased to boot since it only tests “jazzy riffs.”

   That said, I do believe that a person with amusia would find it exceedingly difficult to complete the test, and since that’s what it’s designed to measure, it’s doing it fine.

   One objection: why would only a score of below 50 percent be regarded as “tone deaf”? After all, 50 percent reflects an accuracy level equal to chance. It seems you would need to set the threshold higher than that.

8. #8 dsgoen

   November 13, 2006

   86.1%. But my wife interrupted my listening on two of the sequences.

   I listened via the speakers attached to my computer. Do you think that headphones would have made any difference in the results?

   Jazz bassist in a former life before computers.
   D

9. #9 dsgoen

   November 13, 2006

   Also, I don’t know about other musicians that took this test, but I frequently found myself thinking either that I liked a particular riff or that that I wouldn’t have played that particular sequence. This put me off on something of a tangent, distracting me from the second phrase.

   Most musicians are not soloists, so they are used to interacting with other instrumentalists. This requires a deeper concentration that I generally devote to just listening to music. If I’m concentrating too much on any one aspect of music, then I tend to lose the whole. For example, if I’m playing a call and answer to a sax, I may only have a peripheral consciousness idea of what the piano and drums are doing.

   The more I think about this, the more convinced I am that musicians would score lower than non-musicians that are astute listeners.

10. #10 Jokermage

    November 13, 2006

    That said, I do believe that a person with amusia would find it exceedingly difficult to complete the test, and since that’s what it’s designed to measure, it’s doing it fine.

    I disagree that it is doing fine. People with poor memory could have trouble with this test as well, so it really isn’t distinguishing between tonedeafness and non-tonedeafness.

11. #11 Dave Munger

    November 13, 2006

    “The more I think about this, the more convinced I am that musicians would score lower than non-musicians that are astute listeners.”

    I think you may be right. My musicologist friend says the test is “excruciatingly difficult,” but as a nonmusician, I scored 97.3.

12. #12 Theo Bromine

    November 13, 2006

    I suppose you could remove some of the memory element by allowing the selections to be replayed – presumably this would help people who can distinguish the tones, but would not help tonedeaf people at all. I got 94.4, and there were in fact some that I was unsure of by memory, so I was actually expecting a lower score.

    The sequences seemed to be generated randomly within a few constraints (with the exception of the major scale somewhere in the middle) – maybe that makes them sound like jazz riffs, but I doubt that was intentional. Certainly the more conventionally “tuneful” ones are easier to remember (for me anyway), so perhaps using random tunes is a better way to control for memory effects.

13. #13 Katherine

    November 13, 2006

    hmm…I thought I had done very well but I only scored about 70%. There were a few I couldn’t hear altogether, because some were at such different volumes and I didn’t want to listen to them loudly because there are people around. But I didn’t think that would affect my score much.

    I’m wondering whether I misidentified some as being “different” when really they were the same. I say this because I definitely put “different” more often than I put “same”. Some were ridiculously obviously different from each other. I’m wondering if musicians might have a biased towards saying “different” because some of the phrases contained musical anomolies (at least to the western ear.) So, the second time you hear the phrase, you might think, “that’s wrong” and mark “different” when really it’s musically wrong to the western ear, but is in fact the same as the first phrase.

14. #14 Joe Shelby

    November 13, 2006

    Headphones didn’t help, but I would rather see the actual answers and be able to hear them again. I only got 66.7, answering MOST of them as “different” and my sense of harmony and dissonance and musical memory is very well trained. I am a musician as well as a software developer.

    the nearest i can think is that the *phrases* might be the same but the durations are sometimes different, so the dissonant chord i heard in the second one was not in the first because the first cut off a note sooner. in that, i would call the answer (if “same”) as wrong.

15. #15 Arni

    November 13, 2006

    It would be fun to rip the test off and correlate it to another pitch-perception test, and a auditory memory test and see which correlation would be higher. I thought it was a major strain on my memory, even with headphones and limited external noise.

16. #16 Gav

    November 13, 2006

    I agree it’s a test of memory rather than pitch. I scored 86.1 but suspect that was largely a matter of luck.

17. #17 Theo Bromine

    November 13, 2006

    Agreed that it is a memory challenge, but if the subject can’t detect the pitch differences, it doesn’t matter how well they remember the 2 different sequences.

18. #18 Dave

    November 14, 2006

    I think the main issue with the test is that no matter how good your pitch is, you can’t score very highly unless you also have excellent pitch memory. Thus, it does a good job of picking out people with poor pitch perception, but a not-so-great job of picking our people with good pitch perception, because at the high end pitch perception is conflated with pitch memory.

19. #19 Tony

    November 14, 2006

    I am a musician and my score was 86.1% listening on horrible laptop speakers with some room noise in the background.

    The volume difference between clips made it hard to hear some of them. As someone mentioned earlier, a replay button would have been helpful.

    With my limited attention span, I had a hard time remembering some of the longer clips. I think the test could still judge pitch recognition using shorter clips. This would at least reduce the effects due to memory (or lack thereof).

20. #20 Chris

    November 15, 2006

    I was very surprised that my husband scored exactly the same as I did? 83.3% – I refreshed the screen before he started the test. Now I doubt the validity.

21. #21 Dave Munger

    November 15, 2006

    Chris –

    About 30 percent of respondents were within the 80-90 percent range. Given that there were only four possible scores falling in that range (32/36, 31/36 30/36, and 29/36), chances are actually reasonably good that you and your husband could get the identical score.

22. #22 bioephemera

    November 15, 2006

    I’m not a musician and usually go off-key while singing. In this test, I had no idea what notes were being played, and it would be impossible for me to symbolize them in order to commit them to memory, because I can’t read music. So I just waited to see if I registered a feeling of surprise at the second hearing, and if so, even though I couldn’t say how they differed, I clicked “different.” I got a 75. Since I had no language to encode the information in my memory, I was surprised I did that well!

23. #23 Krytha

    November 17, 2006

    Although the test author says that a replay button would give others a distinct advantage over X,Y and Z, I would disagree. The point of the test is to test tone-deafness is it not? Not how GOOD you are in relation to other people who are NOT tone-deaf. An individual who was tone-deaf could press replay a million times and still not get it. If you’re not tone-deaf I doubt getting a 90% vs. 80% would matter all that much.

24. #24 Chris

    November 22, 2006

    I took this test and another test set up by the governmet. On this test I initially scored around 68% or something like that. The second and third time I took it I got a 75%. I’ve been playing music for 15 years and have a very good ear. I can hear when something just slightly gets out of tune even when my musician friends cannot. I feel that this test just isn’t a good indicator of tone perception. On the government test that I took testing tone perception I got a 100% right. This again, leads me to believe that this test is flawed…