Indeed.
-
WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Can we trust our ears?
- Thread starter mi-fu
- Start date
Indeed.
ernestcarl
Major Contributor
This may sound unfair, but I think when doing these types of tests age should be disclosed and hearing should be tested — as well as testing environment, equipment et al. Whenever doing proper blind studies rigorous controls and methodology must be adhered to... These types of tests we do for ourselves can be sufficient for personal purposes, but if we want hard proof, we have to look at the actual studies that’s already been done.
I can tell you that I do have tinnitus too, but I can still hear up to 16KHz, and most of the differences I notice almost always end up in the high frequencies. But since much of the interpretation of our senses still goes back for processing in the brain, it is really not uncommon for our brains to add all kinds of additional information that really is not there to begin with — e.g. so much of vision is really just filling in the blanks hence ubiquity of optical illusions. I was looking at one ASR member’s 3D family living room plan, and I could have sworn I saw a man sitting on the table on the small image. I zoomed on the image much, more closer and scanned it from corner to corner for a while, and when I finally came back to the spot where I thought I saw a man sitting on the table, turned out it was actually just a poorly 3D rendered flower vase. It took more than a minute for me to realize this! That long... Similar story with the “invisible gorilla” experiment... Our brains LIE to us ALL THE TIME. That’s just a hard fact of life.
I can tell you that I do have tinnitus too, but I can still hear up to 16KHz, and most of the differences I notice almost always end up in the high frequencies. But since much of the interpretation of our senses still goes back for processing in the brain, it is really not uncommon for our brains to add all kinds of additional information that really is not there to begin with — e.g. so much of vision is really just filling in the blanks hence ubiquity of optical illusions. I was looking at one ASR member’s 3D family living room plan, and I could have sworn I saw a man sitting on the table on the small image. I zoomed on the image much, more closer and scanned it from corner to corner for a while, and when I finally came back to the spot where I thought I saw a man sitting on the table, turned out it was actually just a poorly 3D rendered flower vase. It took more than a minute for me to realize this! That long... Similar story with the “invisible gorilla” experiment... Our brains LIE to us ALL THE TIME. That’s just a hard fact of life.
audiophile
Active Member
- Joined
- Oct 7, 2019
- Messages
- 177
- Likes
- 140
Not only our brain can add additional non-existing information to what’s actually there, but it also often removes the information that is in fact exists and may be audible. Why does the music playing in a room sound so different comparing to the same music recorded with a microphone at the listening position? Because the mic captures absolutely everything, including room’s ambience, while our brain filters out the information that seems unimportant.
I think the NPR test could have been way more rigorous in terms of data collection, though they at least ask what basic equipment you're using. It's really just another internet test for entertainment.
Unfortunately, my results (I really wish they kept accessible records) may be of little interest to many. It's certainly of interest to me and, to a lesser extent, my family, as they were the ones who saw it happen. I revealed I could beat the test like I'd uncovered a great conspiracy or blunder. Ta da!. There's likely people who've read my posts and thought "I just don't believe it" and moved on. That's a shame as I am telling the truth and it's not your usual audio forum bullshit about cable transparency or whatever, but lacking hard evidence I don't have anything more to offer than my word. It would be a bloody odd thing to make up too. I could ask for a note from my mum I guess.
As far as my previous test results go, the probability of getting 60 1/3 choices right in a row by chance are astronomically low, way way past the strictest of statistical significance. Once I twigged the lossless sounder "louder" to me there was never any guessing. The reason I posted on this thread was that at the time I just put it down to a flawed test, but I saw no comments to that effect here. Maybe what I interpreted as louder is something else entirely - it seems the only explanation if the tracks are indeed level matched - and that's what interests me. The fact I could tell in a second or two maybe means I was picking up something without really thinking, it was almost instinctual, and that may even be supported by my increasing confusion when listening to more of the track, flicking back and forth a lot and thinking about it. It's now a real curiosity to me that I can't reliably get 6/6 any more. Something, somewhere has changed. First port of call, I'm gonna syringe my ears!.
Unfortunately, my results (I really wish they kept accessible records) may be of little interest to many. It's certainly of interest to me and, to a lesser extent, my family, as they were the ones who saw it happen. I revealed I could beat the test like I'd uncovered a great conspiracy or blunder. Ta da!. There's likely people who've read my posts and thought "I just don't believe it" and moved on. That's a shame as I am telling the truth and it's not your usual audio forum bullshit about cable transparency or whatever, but lacking hard evidence I don't have anything more to offer than my word. It would be a bloody odd thing to make up too. I could ask for a note from my mum I guess.
As far as my previous test results go, the probability of getting 60 1/3 choices right in a row by chance are astronomically low, way way past the strictest of statistical significance. Once I twigged the lossless sounder "louder" to me there was never any guessing. The reason I posted on this thread was that at the time I just put it down to a flawed test, but I saw no comments to that effect here. Maybe what I interpreted as louder is something else entirely - it seems the only explanation if the tracks are indeed level matched - and that's what interests me. The fact I could tell in a second or two maybe means I was picking up something without really thinking, it was almost instinctual, and that may even be supported by my increasing confusion when listening to more of the track, flicking back and forth a lot and thinking about it. It's now a real curiosity to me that I can't reliably get 6/6 any more. Something, somewhere has changed. First port of call, I'm gonna syringe my ears!.
ernestcarl
Major Contributor
No one doubts that you heard what you thought you heard. The fundamental question is whether you can actually trust your brain to provide you with an 1 is to 1 accurate perception of the highest resolution of details found in a recorded piece of music. In many cases the answer is no. That does not make your experience, even if it’s not entirely accurate, meaningless. See: gestalt psychology
Oh I don't know, this is the internet - it's full of wild claims of all sorts. I wouldn't be surprised by (or blame) people who dismissed my account out of hand just because it's statistically improbable.
As to the deeper questions my experience raises, forgive me if I'm surprised that Gestalt psychology is being mentioned - that's the first time I've ever seen it referenced on an audio forum, where you'd think it would have raised its head at some point. Saying that I'm not a frequent visitor to audio forums, probably because I've clearly visited the wrong ones!
I'd like to drill down to the elements that I experienced as increased relative volume - or at least our normal understanding of it. Thinking about it, for most songs I waited until the first large peak in volume and then I moved onto the next track and did the same. It may have been the peak that I interpreted to be louder, more impactful, or any number of other adjectives that are inevitably inadequate in describing sensory perception.
However, in a two second burst of music I'm still listening to the music - god forgive those who hijacked this word - holistically, too. Is it possible my brain was automatically recognising a higher quality file, breaking down the patterns and characteristics for me and the best I could summarise that as was "louder", or "denser"?. Sure, it certainly wasn't conscious. I'm not practiced in 320 kbps mp3 vs lossless vs high res differentiation, I don't have trained ears - I wasn't actively listening for anything except increased volume. In the meantime my brain is doing backflips and spitting out "yes dear, that one's 'louder', you ignorant fool. Moohaha"
As to the deeper questions my experience raises, forgive me if I'm surprised that Gestalt psychology is being mentioned - that's the first time I've ever seen it referenced on an audio forum, where you'd think it would have raised its head at some point. Saying that I'm not a frequent visitor to audio forums, probably because I've clearly visited the wrong ones!
I'd like to drill down to the elements that I experienced as increased relative volume - or at least our normal understanding of it. Thinking about it, for most songs I waited until the first large peak in volume and then I moved onto the next track and did the same. It may have been the peak that I interpreted to be louder, more impactful, or any number of other adjectives that are inevitably inadequate in describing sensory perception.
However, in a two second burst of music I'm still listening to the music - god forgive those who hijacked this word - holistically, too. Is it possible my brain was automatically recognising a higher quality file, breaking down the patterns and characteristics for me and the best I could summarise that as was "louder", or "denser"?. Sure, it certainly wasn't conscious. I'm not practiced in 320 kbps mp3 vs lossless vs high res differentiation, I don't have trained ears - I wasn't actively listening for anything except increased volume. In the meantime my brain is doing backflips and spitting out "yes dear, that one's 'louder', you ignorant fool. Moohaha"
ernestcarl
Major Contributor
As j_j rightly pointed out, we need to get our terminology in check first before we can even begin to discuss what we are actually communicating about with precision.
Loudness vs Volume
Just because the volume increased does not mean we will perceive something as louder.
What is louder to one person may not be louder to another who suffers from some level of hearing loss.
Yes, what you interpreted as louder or quieter may have been something else e.g. compression, filling or clipping of frequency peaks, loss or gain of stereo detail...
Gestalt psychology was very much influential in early studies of perception and perceptual illusions — although, admittedly less so with audio science. You may not have read the exact word or field mentioned in many an audio forums, but some of its influence has carried over quite a bit.
Sorry, was sorting stuff out. In context, when I say I was listening for differences in volume, it's understandable when I express that change as "louder" or "quieter". It could be a useful distinction if, even though the music is apparently the same volume, we're referring to elements of the music that may indeed be higher or lower in volume, or may be interpreted as quieter or louder for whatever reason. If the levels were truly matched and I really do hear lossless as louder then that would suggest in some cases a person's preference for higher res music may be related at least in part to it sounding marginally louder.
This is where my ignorance of lossy codecs comes in. I dont know precisely what could happen to the those music files to make them seem quieter. You mentioned stereo information; tangentially - playing with the crossfeed on the RME is fascinating. The more crossfeed, the quieter it sounds, despite the volume remaining stable. Whether that's a real volume change from the DSP or just psychoacoustics I don't know.
The thing that has me question the validity of any of this is my apparent inability to consistently reproduce the results this morning (although the ear syringe and headphone collection hasn't come out yet!). It wasn't a total flop, but I remember the difference being distinct that once you'd noticed it that was it, so it could even have been a genuine disparity which has since been corrected.
I was speaking to Ms Poo about it earlier; she says it was actually last Christmas (and I was dubious when people said time sped up as you age!) when I was running around shoving headphones on any family member who was even remotely stationary, and seeing how they did. When I later had my revelation on how to cheat the test the only person really interested (and still conscious) was my nephew. I told him to just pick the loudest one and he got them all right a few times in a row. But neither of us could do it from the phone alone. Once he was primed to know what to listen for it became a straightforward task with an amp, although he was already pretty good with his young ears. I'm going to get him to repeat the test using the same technique and see how he fares now. It'll be Xmas by the time I see him but it'll give me time to figure out what gear I was using. Now I'm wondering if a change in phone could possibly have an effect. No. Browser? I'm getting deep into the weeds here, this is daft. I'll try and sort it out this weekend.
I think the last time I literally heard mention of Gestalt psychology was in a philosophy seminar with a psychology student. I mean they get plenty of tourists so fair's fair. He'd taken a philosophy module in epistemology and from what I remember he gave a good showing, though I will never, ever, forgive that ponytail.
This is where my ignorance of lossy codecs comes in. I dont know precisely what could happen to the those music files to make them seem quieter. You mentioned stereo information; tangentially - playing with the crossfeed on the RME is fascinating. The more crossfeed, the quieter it sounds, despite the volume remaining stable. Whether that's a real volume change from the DSP or just psychoacoustics I don't know.
The thing that has me question the validity of any of this is my apparent inability to consistently reproduce the results this morning (although the ear syringe and headphone collection hasn't come out yet!). It wasn't a total flop, but I remember the difference being distinct that once you'd noticed it that was it, so it could even have been a genuine disparity which has since been corrected.
I was speaking to Ms Poo about it earlier; she says it was actually last Christmas (and I was dubious when people said time sped up as you age!) when I was running around shoving headphones on any family member who was even remotely stationary, and seeing how they did. When I later had my revelation on how to cheat the test the only person really interested (and still conscious) was my nephew. I told him to just pick the loudest one and he got them all right a few times in a row. But neither of us could do it from the phone alone. Once he was primed to know what to listen for it became a straightforward task with an amp, although he was already pretty good with his young ears. I'm going to get him to repeat the test using the same technique and see how he fares now. It'll be Xmas by the time I see him but it'll give me time to figure out what gear I was using. Now I'm wondering if a change in phone could possibly have an effect. No. Browser? I'm getting deep into the weeds here, this is daft. I'll try and sort it out this weekend.
I think the last time I literally heard mention of Gestalt psychology was in a philosophy seminar with a psychology student. I mean they get plenty of tourists so fair's fair. He'd taken a philosophy module in epistemology and from what I remember he gave a good showing, though I will never, ever, forgive that ponytail.
ernestcarl
Major Contributor
This student you mentioned must have encountered it in one or more of his courses. I delved into it when I took a course in the history of psychology. My interest in audio science probably is related with my own personal interest in psychophysics and some of the early psychophysical experiments conducted back in the day.
If you have time in the weekend, yeah, you might want read up more on psychoacoustics and perceptual encoding used in lossy compression techniques.
There’s already quite a lot of links to resources here spread around different threads in the forum. But Wikipedia is also good start. Good luck!
If you have time in the weekend, yeah, you might want read up more on psychoacoustics and perceptual encoding used in lossy compression techniques.
There’s already quite a lot of links to resources here spread around different threads in the forum. But Wikipedia is also good start. Good luck!
ernestcarl
Major Contributor
Anyone who wants to learn more about audio perception may want to listen to this interview with j_j from the Home Theater Geeks 8 years ago (only 306 views right now!? It needs more views people!):
I just saw in @pkane signature https://distortaudio.org/, then I found https://www.audiosciencereview.com/...test-distort-audibility-of-distortions.10163/.
Very interesting software, somehow similar with REW - the added harmonics option, and definitely a viable way to test our hearing perception of audio harmonics.
Very interesting software, somehow similar with REW - the added harmonics option, and definitely a viable way to test our hearing perception of audio harmonics.
IndividualWay7
Member
- Joined
- Dec 4, 2019
- Messages
- 37
- Likes
- 21
Oh i never understood golden ear audiophiles who tout they can tell MP3, But never show any proof. Or somehow get offended that AAC, Opus and musepack are transparent at 160kbps even on good gear.
You could try A/B test yourself and see if you can spot any differences. Visually, feel free to use https://deltaw.org/ to check what's changing into the mp3 file over the original WAV/FLAC.
Last edited:
ernestcarl
Major Contributor
So does DeltaWave
I thought DeltaWave was a nulling program. Nulling can tell you if two signals are identical or not. It can't tell if the differences are audible.
- Joined
- Feb 23, 2016
- Messages
- 20,760
- Likes
- 37,614
After you null two files, Deltawave also lets you run an ABX test on yourself using the two files. This is in addition to playing back the difference file from the nulling. I think it is safe to say if the nulled result is inaudible at normal volume you'll not hear a difference. Even then Deltawave lets you add gain to the nulled file so you can push up the level and hear what is left.
That is fine for results of things like distortion or frequency response aberrations. It doesn't really make sense for perceptual encodins like mp3.
Similar threads
- Poll
