• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Relationships between physical sound, auditory sound perception, and music perception

Joined
May 13, 2018
Messages
38
Likes
41
Location
Maryland, USA
#1
Lately, I find myself often pondering how our ears and brains take the physical sound reaching our eardrums and turn that into perceptions of sounds and music. I'm hoping that others in this forum might be interested in discussing this topic, so I'm starting this thread.

I make a distinction between sound and music perception because sound perception need not involve music (a lot of the research in this area is on speech recognition), and music perception involves added layers of interpretation of the sound, emotional responses, judgments of music quality and artist intent, etc.

But I think there are important relationships between sound and music perception, which go in both directions. For example, I'm familiar with the nuances of how some musical instruments sound from playing those instruments myself, which enables me to notice details in how those instruments sound through my gear which I would otherwise not notice, and it also changes my musical perception of those instruments. At the same time, I notice how I bring expectations and moods to music when I listen to it, which changes how I direct my attention in terms of the aspects of the sound I notice (e.g., if I'm focused on a guitar solo, I may hardly notice the tonal character of the drums).

All of this has interesting aspects in terms of what's going on both consciously and subconsciously, and how that all relates to biases, listening experience, training, etc.

Focusing on bias, there doesn't seem to be much professional literature which delves into the specifics of how cognitive biases can affect our auditory perception, but I did run across this recent paper on "Overcoming Bias: Cognitive Control Reduces Susceptibility to Framing Effects in Evaluating Musical Performance."

https://www.nature.com/articles/s41598-018-24528-3

The paper focuses on how bias can affect perception of quality of music performances, rather than perception of sound quality. The basic finding was that, when a listener is told that a performance is by a professional musician rather than a student musician, listeners will tend to perceive the performance as being of higher quality, and this is a more consistent effect than whether the performance was actually by a professional vs student. They used fMRI to see what was happening with people's brains as they listened, and they found that telling a listener that a performer is a professional tended to cause brain responses within a few seconds which would cause the performance to be viewed more favorably, and those brain responses persisted through listening to 70-sec music excerpts, as well as after listening. These brain responses involved the listeners paying more attention. Once a listener was told that a performance was by a professional, it was difficult for listeners to conclude otherwise, even when the performance provided mounting evidence that it wasn't by a professional.

Here are some quotes from the paper:

"By modulating expectations and beliefs, contextual information can alter the enjoyability of stimuli as diverse as artworks, soda, and wine, influencing or even dominating actual sensory perception."

"... contextual information can contribute materially to positive perceptual experiences. Aesthetic experiences sometimes depend on the prior activation of a set of beliefs that dispose a person to perceiving this way—a “preparatory set” consisting of expectations and beliefs. For instance, even though listening to Joshua Bell perform a concert on the violin can cost $100 per ticket, an incognito performance by him at a subway station triggered very little interest. Generally, this evidence suggests that contextual information can affect preferences and perception in both nefarious and beneficial ways."

"Previous neuroimaging studies suggest that the influence of beliefs and expectations arises not merely from the sensory system, but from the particular sensitivity to contextual information of reward structures in the brain."

"Our analysis revealed that, when a piece was preferred, the professional pianist frame induced significantly more activity in the primary auditory cortex relative to the student pianist frame (see Fig. 2A). This suggests that beliefs regarding the quality of a performer engendered a bias in attention."

"We observed higher activation in the primary auditory cortex when the player was described as a professional pianist relative to when the player was described as a student. Moreover, this difference in activity remained consistent, exhibiting no significant changes across the 70 seconds of the excerpt. A panel regression of activity in the primary auditory cortex on time showed no significant linear slope (b1 = 0.0003, z = 0.56, p > 0.5). This supports the notion that a bias in attention began almost immediately (i.e. 4 sec) after the presentation of the framing information and remained stable throughout the evidence accumulation period. Contrary to the notion that more evidence should diminish any framing effects generating during the relatively short framing period (i.e. 4 sec), we found that the professional framing gave rise to a constant attentional bias in favor of the professional player."

"...as information about the quality of the performance accumulated, participants needed to exert cognitive control in order to form and retain a negative evaluation for performances that had been framed as played by a professional compared to those that had been framed as played by a student. These data suggest that less cognitive effort was required to dislike a performance when it had been described as played by a student rather than a professional."

"... by expecting better performance from a professional, participants directed more attention toward professionally framed pianists compared to the student-framed performances, and therefore, exhibited a heightened tendency to gather more evidence that would confirm their prior expectation about the professional player’s performance."

"From the perspective of music psychology, these findings reinforce the notion that extrinsic factors—outside the borders of the notes themselves—can impact perception and evaluation as critically as the intrinsic characteristics of the acoustic signal."
 
Joined
Dec 1, 2018
Messages
39
Likes
28
#2
I would like to have seen whether, if you had a pool of true expert musicians evaluating the performances, the cognitive bias as to what was a good performance versus a poor one, based on the labelling of the performer as "professional" or not, would have disappeared. In other words, I hypothesize the expert professional musicians would have made better judgments.

I read the other day, oh, I don't know where, that perhaps one in 50,000 people has what it takes to be a professional musician. I hypothesize that those among the 1-in-50,000 will have the skills and confidence to judge more independently.

Conversely, if one is not in that rarefied air, in the 1 in 50,000, he has less to bring to the table in evaluating a performance. He looks for and hopes to trust extraneous clues.

Does that make sense? Is it at all toward your point?

Tangentially, this also probably accounts for at least some of musicians' contempt for professional critics of their music.
 
Last edited:

Wombat

Major Contributor
Joined
Nov 5, 2017
Messages
2,886
Likes
1,179
Location
Australia
#3
Let us condense this down to a reasoned and scientific level and leave the multiple conjectures to be open to credible discourse.

This forum continues to drift off-course. :facepalm:
 
Joined
Jul 21, 2018
Messages
485
Likes
638
Location
The Neverlands
#4
Condensing....

Relationships between physical sound, auditory sound perception, and music perception

Looks like a personal thing to me given the words perception.

I don't think it really possible to get an 'average' value for this.
For each of us there is a clear relationship though.

A crappy, or crappy reproduced recording (the physical sound) will most likely sound poor to each of us who care about sound quality.
The 'who cares about sound quality' is important because the vast majority of people on this planet really don't care at all.

Well made (and reproduced) recordings are usually 'well perceived' by all people.
Most people I met that do not have any affinity with sound quality all mention that it sounds 'great' in that case.
The difference is they don't want to pay the money needed for this but they do hear it.

So yeah, there seems to be a clear relation between physical sound and perception.
 

Wombat

Major Contributor
Joined
Nov 5, 2017
Messages
2,886
Likes
1,179
Location
Australia
#5
Condensing....

Relationships between physical sound, auditory sound perception, and music perception

Looks like a personal thing to me given the words perception.

I don't think it really possible to get an 'average' value for this.
For each of us there is a clear relationship though.

A crappy, or crappy reproduced recording (the physical sound) will most likely sound poor to each of us who care about sound quality.
The 'who cares about sound quality' is important because the vast majority of people on this planet really don't care at all.

Well made (and reproduced) recordings are usually 'well perceived' by all people.
Most people I met that do not have any affinity with sound quality all mention that it sounds 'great' in that case.
The difference is they don't want to pay the money needed for this but they do hear it.

So yeah, there seems to be a clear relation between physical sound and perception.
Enjoyment of music transcends modern constructs re sound quality, except for a narrow focussed minority. ;)

Better sound quality is nice but not an absolute requirement.
 

Cosmik

Major Contributor
Joined
Apr 24, 2016
Messages
2,314
Likes
1,029
Location
UK
#6
"From the perspective of music psychology, these findings reinforce the notion that extrinsic factors—outside the borders of the notes themselves—can impact perception and evaluation as critically as the intrinsic characteristics of the acoustic signal."
So even if a test is double blind, the assumptions made by the subjects of why they are taking part, or of what they might be listening to, can affect the outcome. If you tell the subjects what they are going to hear, you affect the outcome, and if you don't tell them what they are going to hear they make their own assumptions that affect the outcome. And that "effect on the outcome" can literally be to reverse the finding.

It goes back to the discussions on the nature of science which usually boil down to: if you report accurately what you did and don't draw unwarranted conclusions, it is science.

Logically, then, if we accept the first paragraph, the only conclusion that is warranted from any listening test is "This is what we did and this is what we found." and nothing else. As soon as you add "Therefore we feel justified in claiming that our wider hypothesis is confirmed: second harmonic distortion improves perception of an audio system." you have overshot what is warranted.

Therefore the science doesn't help you improve your audio system: you can't acknowledge that a listening test's results are inextricably linked with the subjects' assumptions and mental states but then take the results and use them to justify changing/not changing your hardware when the hardware is going to be used by people with different assumptions and mental states from those of the test's subjects.

But this is exactly what people do. The same people who, when pressured, invoke the "Science just means you report it accurately" clause, also use the "Science says humans don't hear phase distortion" assertion and use it to justify designing non-neutral hardware.
 
Last edited:
Joined
May 13, 2018
Messages
38
Likes
41
Location
Maryland, USA
#7
Let us condense this down to a reasoned and scientific level and leave the multiple conjectures to be open to credible discourse.

This forum continues to drift off-course. :facepalm:
Please do focus on anything particular you'd like to discuss. Per the thread title, the aim of this thread is explore the connections between three areas - physical sound, sound perception, and music perception - whereas most psychoacoustic audio discussions focus on only physical sound and sound perception. Since we're all using audio gear specifically to listen to music (rather than speech, etc.), I think there's value in bringing in the music perception aspect.

In that regard, one thought I can add to original post is the sometimes 'antagonistic' relationship between sound perception and music perception. While it seems natural to expect that better sound quality (let's say 'higher fidelity', though that's a topic in itself) leads to enhanced music appreciation, I do think many audiophiles often focus their attention so much on sound quality (e.g., 'how realistic do the cymbals sound?') that it can take attention away from the higher-level music content and thus diminish music appreciation. I'm not saying that we can't simultaneously pay attention to both sound quality and music, but we do have limited mental bandwidth, so there may be a sort of uncertainty principle involved where attention to one comes at some expense to the other.
 
Last edited:
Joined
May 13, 2018
Messages
38
Likes
41
Location
Maryland, USA
#8
I would like to have seen whether, if you had a pool of true expert musicians evaluating the performances, the cognitive bias as to what was a good performance versus a poor one, based on the labelling of the performer as "professional" or not, would have disappeared. In other words, I hypothesize the expert professional musicians would have made better judgments.

I read the other day, oh, I don't know where, that perhaps one in 50,000 people has what it takes to be a professional musician. I hypothesize that those among the 1-in-50,000 will have the skills and confidence to judge more independently.

Conversely, if one is not in that rarefied air, in the 1 in 50,000, he has less to bring to the table in evaluating a performance. He looks for and hopes to trust extraneous clues.

Does that make sense? Is it at all toward your point?

Tangentially, this also probably accounts for at least some of musicians' contempt for professional critics of their music.
Yes, I agree. I think that what we perceive, whether related to music or sound, is very much dependent on the 'top-down' framework we bring to the act of perception, in terms of our prior experiences, beliefs, expertise, emotional associations, mood, energy level, social context, expectations, heuristics and biases, etc.

To extend your experiment, it would be interesting to see how expert musicians from different cultures would have evaluated the performances.
 
Joined
May 13, 2018
Messages
38
Likes
41
Location
Maryland, USA
#9
So even if a test is double blind, the assumptions made by the subjects of why they are taking part, or of what they might be listening to, can affect the outcome. If you tell the subjects what they are going to hear, you affect the outcome, and if you don't tell them what they are going to hear they make their own assumptions that affect the outcome. And that "effect on the outcome" can literally be to reverse the finding.

It goes back to the discussions on the nature of science which usually boil down to: if you report accurately what you did and don't draw unwarranted conclusions, it is science.

Logically, then, if we accept the first paragraph, the only conclusion that is warranted from any listening test is "This is what we did and this is what we found." and nothing else. As soon as you add "Therefore we feel justified in claiming that our wider hypothesis is confirmed: second harmonic distortion improves perception of an audio system." you have overshot what is warranted.

Therefore the science doesn't help you improve your audio system: you can't acknowledge that a listening test's results are inextricably linked with the subjects' assumptions and mental states but then take the results and use them to justify changing/not changing your hardware when the hardware is going to be used by people with different assumptions and mental states from those of the test's subjects.

But this is exactly what people do. The same people who, when pressured, invoke the "Science just means you report it accurately" clause, also use the "Science says humans don't hear phase distortion" assertion and use it to justify designing non-neutral hardware.
I think it's fair to say that science typically does involve inference and generalization from the finite and context-specific findings of particular studies. It involves judgment (hence opening the door to debates) and entails risk of various types of errors, but I don't see any way around it. We're unavoidably swimming in uncertainty, extrapolating and interpolating from the known to unknown, hoping that it works out. If we do lots of studies and look at things from lots of angles and triangulate all of that, we generally have a better chance of things working out - but that itself is a generalization!
 

andreasmaaan

Major Contributor
Patreon Donor
Joined
Jun 19, 2018
Messages
1,801
Likes
1,049
#10
Interesting post @Phronesis. I guess it shouldn’t be any surprise that in our perception of musical quality, as in all matters of perception (including of audio), expectation bias may play a role.
 
Joined
Dec 1, 2018
Messages
39
Likes
28
#11
Interesting post @Phronesis. I guess it shouldn’t be any surprise that in our perception of musical quality, as in all matters of perception (including of audio), expectation bias may play a role.
I am concerned however, that the 40 participants in the study had no formal musical training. Perhaps that did control for the variable of amount of musical training, and from what I have read I over-estimate how many people have formal musical training (it is a fraction of what I would have expected), but still, the study asks people to engage in higher-order musical attention (as I think I remember one poster referencing it) but with no formal musical training, and then judge the quality of music. That just hits me wrong. The words that come to my head is it's a train wreck waiting to happen.

One thing I found super-interesting was the observation in this thread is that paying attention to audio quality can distract away from higher-order musical listening. That seems like a very astute observation. When people upgrade to new equipment and expect to hear details they never heard before, if it is not too dramatic an upgrade (or arguably of a placebo-like variety), I suggest to them if they listen closely to a piece of music three or four times on their existing equipment I 100 percent guarantee them they will hear details they never heard before, and at a much lower cost in time and money, if not effort. It does require mental effort, it is not handed to you on a silver platter.

Another thing I am getting from this is plasticity of hearing perception--that your ears adapt to the sound and sound quality of the system. We here (including myself) do have an interest of sound quality of the system, but when I upgrade, I ask, what am I getting here that my ears would not have adapted to in the first place? And then I read that people will go back to the old setup and impute the new characteristics of the upgraded system into their old system!

So me personally, I recently added a second subwoofer and felt like I got a visceral improvement in bass quality and uniformity, and I recently bought a nicer receiver with room EQ and manual EQ and baked-in music services, and I felt like I was getting a viscerally ascertainable improvement of mid-range reproduction and enjoyed greatly the new manual control I had over the sound with an automated process as a baseline, and easier access to my music by reason of the features of the receiver. This improved my "music perception" experience.

But there is always that nagging doubt--would I have enjoyed the music just the same had I not upgraded? I can't level match and double-blind these complex changes I have made and as we've seen expectation bias just pops out of nowhere all over the place in music performance, reproduction, and judgment of quality of sound and performance! It never goes away, even if we're double-blind testing! So, even having used my best judgment and information, I am left in a state of some doubt. As the original poster pointed out elsewhere (or at least to paraphrase), to be in a state of doubt is uncomfortable, but to be in a state of certainty is absurd. I believe this was an observation of Voltaire--most likely paraphrased poorly by yours truly--I neither translate from French nor have a perfect memory. :)
 
Last edited:
Joined
May 13, 2018
Messages
38
Likes
41
Location
Maryland, USA
#12
I don't have the full paper, but this is interesting. It shows how, when we listen to exactly the same thing more than once, we can perceive it as being different, and that can be influenced by expectations.

http://mp.ucpress.edu/content/35/1/94

"The repeated recording illusion refers to the phenomenon in which listeners believe to hear different musical stimuli while they are in fact identical. The present paper aims to construct an experimental paradigm to enable the systematic measurement of this phenomenon, investigating potentially related extrinsic and individual difference factors. Participants were told to listen to “different” musical performances of an original piece when in fact they were exposed to the same repeated recording. Each time, the recording was accompanied by a text suggesting a low, medium, or high prestige of the performer. Most participants (75%) believed that they had heard different musical performances. Participants with high levels of neuroticism and openness were significantly more likely to fall for the illusion. While the explicit information presented with the music influenced participants’ ratings significantly, the effect of repeated exposure was only significant in the more familiar music condition. These results suggest that like many other human judgments, evaluations of music also rely on cognitive biases and heuristics that do not depend on the stimuli themselves. The repeated recording illusion can constitute a useful paradigm for investigating nonmusical factors because it allows for the study of their effects while the music remains the same."​
 
Last edited:

Don Hills

Addicted to Fun and Learning
Joined
Mar 1, 2016
Messages
568
Likes
185
Location
Wellington, New Zealand
#13
My primary criterion for judging the quality of a piece of music (or other art) is how well it rewards repeated experiencing with new insights. My secondary criterion is the demonstrated expertise of the artist. When they are simultaneously present in a piece, the effect is magical.
 
Joined
May 13, 2018
Messages
38
Likes
41
Location
Maryland, USA
#14
My primary criterion for judging the quality of a piece of music (or other art) is how well it rewards repeated experiencing with new insights. My secondary criterion is the demonstrated expertise of the artist. When they are simultaneously present in a piece, the effect is magical.
… which leads me to note that, when I was young, I could listen to a piece 100+ times and find that my enjoyment of it didn't diminish at all, and would typically increase for a while before possibly reaching a plateau. But now that I'm older (over 50), I find that I "use up" a piece of music in a lot fewer listens, and feel greater need to hear new music (which Spotify and Tidal are great for). How do everyone else's experiences compare in this regard? (I guess this is sort of relevant to the thread topic - changing perceptions of the same thing over time.)
 
Joined
Dec 1, 2018
Messages
39
Likes
28
#15
… which leads me to note that, when I was young, I could listen to a piece 100+ times and find that my enjoyment of it didn't diminish at all, and would typically increase for a while before possibly reaching a plateau. But now that I'm older (over 50), I find that I "use up" a piece of music in a lot fewer listens, and feel greater need to hear new music (which Spotify and Tidal are great for). How do everyone else's experiences compare in this regard? (I guess this is sort of relevant to the thread topic - changing perceptions of the same thing over time.)
If I really like a piece of music and it is new to me I can listen to it 5 or 10 or 20 times and get a real super-high off the experience. I don’t really have time in life to hit the plateau the way I did as a teenager. I feel like my mind opened back up at some point in mid-adulthood. I walked into a guitar store not having played for years and picked up a guitar and a guy sat down next to me and made soft suggestions on how to play better. Slow it down, keep the beat, etc. I was in only barely hidden tears and I felt my heart and my creativity open back up. I then spent a few years bypassing my kid self on the piano. And, I think this is what got me to where I am today, where a piece of music can get the endorphins flying on repeated listening even in my older years. More than you wanted to know I am sure!

It reminds me of what someone told me when I started a new job and was in over my head. A peer talked to me about his experience learning Spanish as an adult, living with an Hispanic family, and he overheard the child ask the mother in Spanish if the man was stupid. And he did learn Spanish, but put himself back into the humble place of a child to learn it. And that is what I had to do to catch up and pass my younger self in so many things after I had experienced the hardening of adulthood—to soften up and learn as a child. And this includes opening the door to new music again and enjoying it as I would have encountered it as a child. What is painful to get there for me is reliving the humility of childhood. To learn again as a child—it opens so many doors, and ironically, it greatly sharpens the wits.
 
Last edited:

Wombat

Major Contributor
Joined
Nov 5, 2017
Messages
2,886
Likes
1,179
Location
Australia
#16
I am concerned however, that the 40 participants in the study had no formal musical training. Perhaps that did control for the variable of amount of musical training, and from what I have read I over-estimate how many people have formal musical training (it is a fraction of what I would have expected), but still, the study asks people to engage in higher-order musical attention (as I think I remember one poster referencing it) but with no formal musical training, and then judge the quality of music. That just hits me wrong. The words that come to my head is it's a train wreck waiting to happen.

One thing I found super-interesting was the observation in this thread is that paying attention to audio quality can distract away from higher-order musical listening. That seems like a very astute observation. When people upgrade to new equipment and expect to hear details they never heard before, if it is not too dramatic an upgrade (or arguably of a placebo-like variety), I suggest to them if they listen closely to a piece of music three or four times on their existing equipment I 100 percent guarantee them they will hear details they never heard before, and at a much lower cost in time and money, if not effort. It does require mental effort, it is not handed to you on a silver platter.

Another thing I am getting from this is plasticity of hearing perception--that your ears adapt to the sound and sound quality of the system. We here (including myself) do have an interest of sound quality of the system, but when I upgrade, I ask, what am I getting here that my ears would not have adapted to in the first place? And then I read that people will go back to the old setup and impute the new characteristics of the upgraded system into their old system!

So me personally, I recently added a second subwoofer and felt like I got a visceral improvement in bass quality and uniformity, and I recently bought a nicer receiver with room EQ and manual EQ and baked-in music services, and I felt like I was getting a viscerally ascertainable improvement of mid-range reproduction and enjoyed greatly the new manual control I had over the sound with an automated process as a baseline, and easier access to my music by reason of the features of the receiver. This improved my "music perception" experience.

But there is always that nagging doubt--would I have enjoyed the music just the same had I not upgraded? I can't level match and double-blind these complex changes I have made and as we've seen expectation bias just pops out of nowhere all over the place in music performance, reproduction, and judgment of quality of sound and performance! It never goes away, even if we're double-blind testing! So, even having used my best judgment and information, I am left in a state of some doubt. As the original poster pointed out elsewhere (or at least to paraphrase), to be in a state of doubt is uncomfortable, but to be in a state of certainty is absurd. I believe this was an observation of Voltaire--most likely paraphrased poorly by yours truly--I neither translate from French nor have a perfect memory. :)
Musical training is a fraught term - general, specialised. analytic, expertise, et al. Given that most audiophiles are probably somewhat tone-deaf in the proper sense then using 'ordinary Joes' as test subjects for typical consumer set-ups is probably more telling for practicability. IMHO.
 
Joined
May 13, 2018
Messages
38
Likes
41
Location
Maryland, USA
#17
Musical training is a fraught term - general, specialised. analytic, expertise, et al. Given that most audiophiles are probably somewhat tone-deaf in the proper sense then using 'ordinary Joes' as test subjects for typical consumer set-ups is probably more telling for practicability. IMHO.
I tend to agree. Though, unfortunately, the paper doesn't clearly spell out their criteria for "without formal training in music." Someone could obviously be a self-taught accomplished musician, despite not having "formal training," and would therefore be a lot closer to a professional musician than a non-musician.
 
Joined
May 13, 2018
Messages
38
Likes
41
Location
Maryland, USA
#19
Off-topic slightly perhaps, but apart from the expectation bias aspect, this study is also consistent with research into the effects of authority on human behaviour, eg Milgram’s classic experiment.
Yes, one of the most famous studies in social psychology. It deals more with behavior than perception, but there are plenty of studies demonstrating how our perceptions can be socially influenced (and salesmen rely on this!).

When I was younger, a friend who was an orchestral percussionist told me that auditions for positions in orchestras are often done blinded (behind screens) so that the performance of candidates can be evaluated without the biases resulting from knowing who the candidate is, their appearance, gender, etc.
 
Last edited:
Top Bottom