5 ms·

> Deutsch posed a question that really got me thinking: how come most people can identify and label colors with ease without needing any “reference colors” to d

> Deutsch posed a question that really got me thinking: how come most people can identify and label colors with ease without needing any “reference colors” to do so, but absolute pitch is so rare People usually identify and label colors with much lower granularity. There are 7 conventional spectral colors with wide subjective boundaries between them, but e.g. piano has 88 distinct tones and that is still not full audible range.

tzs 3y ago
What are the limits of color discrimination, I wonder? Can we discriminate between close colors of light better than we can between close frequencies of sound? For pitch Googling suggests we can hear a difference of about 5 cents. A cent is 1/100th of a semitone, or 1/1200th of an octave, so N cents is a frequency ratio of 2^(N/1200). At 440 Hz (A above middle C) 5 cents would be a difference of 1.27 Hz. For light, a 5 cents difference would be about 2 nm wavelength at the red end of our visible range and about 1.1 nm at other end. (That's about 1 gHz frequency difference at the red and and 2 gHz at the other end). Can a human tell if two light sources differ in wavelength by 1-2 nm?
- pclmulqdq 3y ago
  A trained musician can hear a pitch difference much smaller than 5 cents. I think designers and artists can also likely see a difference in wavelength of 1-2 nm, and probably a lot smaller.
  - mrob 3y ago
    Are they actually hearing the pitch difference, or are they hearing beat frequencies between simultaneous notes? You can hear beat frequencies even in monophonic music if the reverb is heavy enough.
    - pclmulqdq 3y ago
      Pitch difference is what I am referring to. Play one note after the other, and many musicians can hear 1 cent up or down. With simultaneous notes, counting the beats you can hear much smaller differences.
- itishappy 3y ago
  Yup, think that's right about the limit of the eye. It's neat that spectral resolution is so similar for both eyes and ears, particularly considering how different their range is: Ears: about 10 octaves Eyes: about 1 octave
  - LoganDark 3y ago
    > It's neat that spectral resolution is so similar for both eyes and ears, particularly considering how different their range is Probably because it's not necessarily just resolution of the eyes and ears, but resolution of the brain that perceives that information. How "perception" works is still somewhat of a mystery.
GloriousKoji 3y ago
I got this board game called "Hues and Cues"; it has a large board with a color grid and players take turns drawing cards of a color on the board and give a couple verbal hints to try to get the other players to guess the color. After playing it with multiple people and adjusting the rules all sorts of ways I don't think humans are as good as identifying colors as you think.
pclmulqdq 3y ago
Previously, I had heard that the best way of teaching kids absolute pitch is to have them listen to modern jazz and modern classical music, where unexpected pitches often show up. That innately trains them to think of individual note pitch as an important detail to listen to. That would suggest that reference pitches aren't needed, beyond giving each note a name later in life when the kids go for musical training, which in turn means that color and pitch discrimination are actually fairly similar processes, we just decide that one is important and the other is not. I didn't realize there were methods like Deutsch's, and they honestly seem a little wonky. I'm glad it worked for her.
- nescioquid 3y ago
  > ... listen to modern jazz and modern classical music, where unexpected pitches often show up. That innately trains them to think of individual note pitch as an important detail to listen to. I don't think the absolute value any pitch class carries is of much intrinsic musical interest. Musical comprehension is not aided by knowing the pitch-class of a specific note, but in perceiving its relationship to other tones. I'm guessing you picked those two genres of music because they are often referred to as "dissonant" or "noise". As far as modern classical music is concerned, much of the material has a unity between the foreground, background, and middle-ground details. Imagine you've been listening to an orchestra busily sawing away on pitch set [0 1 4] (in all its permutations) and then you realize the lonely glockenspiel's three disembodied notes over the last minute and a half spelled [0 1 4]. Training someone to listen to such a piece so that they might tell you the glockenspiel's second note was an e-flat does not improve anyone's musical comprehension. Training someone to listen to such a piece so that pitch-relationships in the foreground can be heard to form relationships at different scales of the composition does greatly aid in musical comprehension (and is routinely accomplished -- pretty much anyone can do it).
  - pclmulqdq 3y ago
    I think the big thing to note here is that kids don't intuitively do Schenkerian analysis on the music they listen to. I agree with you that a lot of music has a lot more internal structure than most people think, but it's reductive to say that dissonant music contains only "expected" pitches (as heard by an untrained ear).

tzs 3y ago

What are the limits of color discrimination, I wonder? Can we discriminate between close colors of light better than we can between close frequencies of sound? For pitch Googling suggests we can hear a difference of about 5 cents. A cent is 1/100th of a semitone, or 1/1200th of an octave, so N cents is a frequency ratio of 2^(N/1200). At 440 Hz (A above middle C) 5 cents would be a difference of 1.27 Hz. For light, a 5 cents difference would be about 2 nm wavelength at the red end of our visible range and about 1.1 nm at other end. (That's about 1 gHz frequency difference at the red and and 2 gHz at the other end). Can a human tell if two light sources differ in wavelength by 1-2 nm?

pclmulqdq 3y ago
A trained musician can hear a pitch difference much smaller than 5 cents. I think designers and artists can also likely see a difference in wavelength of 1-2 nm, and probably a lot smaller.
- mrob 3y ago
  Are they actually hearing the pitch difference, or are they hearing beat frequencies between simultaneous notes? You can hear beat frequencies even in monophonic music if the reverb is heavy enough.
  - pclmulqdq 3y ago
    Pitch difference is what I am referring to. Play one note after the other, and many musicians can hear 1 cent up or down. With simultaneous notes, counting the beats you can hear much smaller differences.
itishappy 3y ago
Yup, think that's right about the limit of the eye. It's neat that spectral resolution is so similar for both eyes and ears, particularly considering how different their range is: Ears: about 10 octaves Eyes: about 1 octave
- LoganDark 3y ago
  > It's neat that spectral resolution is so similar for both eyes and ears, particularly considering how different their range is Probably because it's not necessarily just resolution of the eyes and ears, but resolution of the brain that perceives that information. How "perception" works is still somewhat of a mystery.

GloriousKoji 3y ago

I got this board game called "Hues and Cues"; it has a large board with a color grid and players take turns drawing cards of a color on the board and give a couple verbal hints to try to get the other players to guess the color. After playing it with multiple people and adjusting the rules all sorts of ways I don't think humans are as good as identifying colors as you think.

pclmulqdq 3y ago

Previously, I had heard that the best way of teaching kids absolute pitch is to have them listen to modern jazz and modern classical music, where unexpected pitches often show up. That innately trains them to think of individual note pitch as an important detail to listen to. That would suggest that reference pitches aren't needed, beyond giving each note a name later in life when the kids go for musical training, which in turn means that color and pitch discrimination are actually fairly similar processes, we just decide that one is important and the other is not. I didn't realize there were methods like Deutsch's, and they honestly seem a little wonky. I'm glad it worked for her.

nescioquid 3y ago
> ... listen to modern jazz and modern classical music, where unexpected pitches often show up. That innately trains them to think of individual note pitch as an important detail to listen to. I don't think the absolute value any pitch class carries is of much intrinsic musical interest. Musical comprehension is not aided by knowing the pitch-class of a specific note, but in perceiving its relationship to other tones. I'm guessing you picked those two genres of music because they are often referred to as "dissonant" or "noise". As far as modern classical music is concerned, much of the material has a unity between the foreground, background, and middle-ground details. Imagine you've been listening to an orchestra busily sawing away on pitch set [0 1 4] (in all its permutations) and then you realize the lonely glockenspiel's three disembodied notes over the last minute and a half spelled [0 1 4]. Training someone to listen to such a piece so that they might tell you the glockenspiel's second note was an e-flat does not improve anyone's musical comprehension. Training someone to listen to such a piece so that pitch-relationships in the foreground can be heard to form relationships at different scales of the composition does greatly aid in musical comprehension (and is routinely accomplished -- pretty much anyone can do it).
- pclmulqdq 3y ago
  I think the big thing to note here is that kids don't intuitively do Schenkerian analysis on the music they listen to. I agree with you that a lot of music has a lot more internal structure than most people think, but it's reductive to say that dissonant music contains only "expected" pitches (as heard by an untrained ear).