Why are we so confident of long term listening evaluations?

[esldude]

Why We Remember So Many Things Wrong - The New Yorker

 

Interesting food for thought on this topic.

 

Shorter listening comparisons are decried as not as good as long term listening evaluations. Yet how reliable are one's memories long term? How pliable are they to being corrupted by other factors?

 

A short excerpt from the article:

 

Their memories were vivid, clear—and wrong. There was no relationship at all between confidence and accuracy.

[mayhem13]

I think SandyK (Alex) said it best in another thread without the direct quote but he suggested that enlisting emotions in a listening evaluation along with longer sessions was a more accurate evaluation than ABX or DBT.

 

I would only submit that once an emotional response has been established, i'm not sure a reliable evaluation can be attained.

 

I have never been able to convince myself of a significant aural memory ability for myself. I CAN isolate artifacts within a piece of complex music and look to identify them repeatedly and with different gear and configurations. Periods of time have NO EFFECT on this identification process and it has proven to be extremely reliable for me. I would add that the fundamental needs to be in the 1khz-5khz range where our ears are most sensitive.

[wgscott]

If we didn't have the gold standard of long-term listening evaluations, how would we have ever been able to discover that double-blind testing doesn't work?

[christopher3393]

If we didn't have the gold standard of long-term listening evaluations, how would we have ever been able to discover that double-blind testing doesn't work?

 

Sarcasm! Soooo helpful!

[Allan F]

Sarcasm! Soooo helpful!

 

It's Bill Scott! What did you expect, honey coated platitudes?

[Jud]

Why We Remember So Many Things Wrong - The New Yorker

 

Interesting food for thought on this topic.

 

Shorter listening comparisons are decried as not as good as long term listening evaluations. Yet how reliable are one's memories long term? How pliable are they to being corrupted by other factors?

 

A short excerpt from the article:

 

Their memories were vivid, clear—and wrong. There was no relationship at all between confidence and accuracy.

 

I agree that *comparisons* over longer periods of time, particularly more detailed ones, are quite unreliable.

[christopher3393]

Simply suggesting it might not be helpful if we want to discuss this.

 

Here's another quote:

memories of emotional events do indeed differ substantially from regular memories. When it comes to the central details of the event, they are clearer and more accurate. But when it comes to peripheral details, they are worse. And our confidence in them, while almost always strong, is often misplaced

 

Now, how do we apply this to audio listening over time? Usually this doesn't mean over years, but over months with repeated exposure. Also not sure if it is usually what we would call heightened emotional memory like some of the incidents used in this study.

 

Even if it is this audio memory is heightened emotional memory, not clear what would be meant by central detail memory, which is improved, and peripheral detail memory which is worse, is applied to listening to a system to evaluate sound. Maybe I forget what I ate or what kind of day it was, but my central focus is on the sound and its characteristics. So not sure that musical detail would be peripheral here.

 

It is interesting though, but not easy for me to make an analogy with listening evaluation. What do you think?

 

maybe we're talking about apples and oranges here, but some reviewers, including Chris C., find it important to listen to a review component over a period of a month or 2. Forgive my ignorance. What am I missing?

 

I'm the honey-coated platitudes guy

[Allan F]

It is interesting though, but not easy for me to make an analogy with listening evaluation. What do you think?

 

I also have reservations about the application of this article to audio. If audio memory is as fleeting as suggested by those who argue that its short duration makes it unreliable, why can I hear the first two notes of a very familiar recording and immediately recognize it, although I haven't listened to it for months or even longer? Are there different factors operating here? And if so, what are they?

[StephenJK]

I'm not sure that confidence has much to do with it, maybe more the comfort of familiarity.

 

We want a system that's capable of playing back the music more or less the way it was recorded.

 

i think that a lot of us take this much too far in the wrong direction. A Clearaudio turntable that sells for over $100K isn't the furthest extrapolation of the "art of the turntable", it's a fashion statement.

 

Some/Most/None of us would agree to this: Listen to live music whenever possible. Put together a stereo system that plays music in a way that you are comfortable with. If and when you reach that point, relax and buy more music.

 

Some of it will be very well recorded and will sound great. Most of it will not, but it's the music that matters. And if it's poorly recorded, then so be it.

[Allan F]

i think that a lot of us take this much too far in the wrong direction. A Clearaudio turntable that sells for over $100K isn't the furthest extrapolation of the "art of the turntable", it's a fashion statement.

 

You wouldn't want to tell that to Mikey Fremer about his $175,000 Continuum Audio Labs Caliburn turntable. Not unless you are prepared for a very heated argument. And that's not counting the stand, which adds another $25,000.

[esldude]

I think SandyK (Alex) said it best in another thread without the direct quote but he suggested that enlisting emotions in a listening evaluation along with longer sessions was a more accurate evaluation than ABX or DBT.

 

I would only submit that once an emotional response has been established, i'm not sure a reliable evaluation can be attained.

 

I have never been able to convince myself of a significant aural memory ability for myself. I CAN isolate artifacts within a piece of complex music and look to identify them repeatedly and with different gear and configurations. Periods of time have NO EFFECT on this identification process and it has proven to be extremely reliable for me. I would add that the fundamental needs to be in the 1khz-5khz range where our ears are most sensitive.

 

This describes very well how evaluations are for me as well. You and Alex together covered the situation succinctly.

[StephenJK]

You wouldn't want to tell that to Mikey Fremer about his $175,000 Continuum Audio Labs Caliburn turntable. Not unless you are prepared for a very heated argument. And that's not counting the stand, which adds another $25,000.

 

I would agree with this: It's important if you want it to be.

[mayhem13]

Sarcasm! Soooo helpful!

 

I think what Bill wishes to express is a stark contradiction that if one disagrees with the premise of Dennis' post becomes sarcastic in nature. Perspective is key here IMO.

 

Would you agree the contradiction of the efficacy of both exists?

[mayhem13]

We want a system that's capable of playing back the music more or less the way it was recorded.

 

 

And how would we know the 'way' it was recorded?...........What's the reference?

 

If Audio and the audiophile hobby is of a subjective nature, all reference to anything is void, including 'Hi Fidelity'.

[Allan F]

And how would we know the 'way' it was recorded?...........What's the reference?

 

Please refer to several threads about "the absolute sound" concept for the answer.

 

If Audio and the audiophile hobby is of a subjective nature, all reference to anything is void, including 'Hi Fidelity'.

 

No, it's not.

[Jud]

If audio memory is as fleeting as suggested by those who argue that its short duration makes it unreliable, why can I hear the first two notes of a very familiar recording and immediately recognize it, although I haven't listened to it for months or even longer? Are there different factors operating here? And if so, what are they?

 

Some very good questions. Stuff I hope to explore over in the "Here's a little test..." thread after folks are done testing and we get into the conversation part of things: www.computeraudiophile.com/f8-general-forum/heres-little-test-you-can-all-join-23330/

 

(Perhaps Dennis started this long term listening eval thread to discuss similar sorts of things.)

[esldude]

Please refer to several threads about "the absolute sound" concept for the answer.

 

 

 

No, it's not.

 

That is Mayhem's point. The Absolute Sound concept has lots of holes in it.

 

Do a little recording of music and you will quickly come to see how damaging that is to the idea of some subjectively discernible TAS comparison using subjectivity which is anchorless.

[esldude]

Some very good questions. Stuff I hope to explore over in the "Here's a little test..." thread after folks are done testing and we get into the conversation part of things: www.computeraudiophile.com/f8-general-forum/heres-little-test-you-can-all-join-23330/

 

(Perhaps Dennis started this long term listening eval thread to discuss similar sorts of things.)

 

Similar concepts yes. Wasn't pre-empting your coming discussion from that test thread. Happened to read the article and it fit well enough to post about.

[Allan F]

That is Mayhem's point. The Absolute Sound concept has lots of holes in it.

 

Do a little recording of music and you will quickly come to see how damaging that is to the idea of some subjectively discernible TAS comparison using subjectivity which is anchorless.

 

At the risk of offending Bill Scott, I would refer to the extensive experience of Barry Diament, which I believe contradicts your suggestion.

[StephenJK]

And how would we know the 'way' it was recorded?...........What's the reference?

 

If Audio and the audiophile hobby is of a subjective nature, all reference to anything is void, including 'Hi Fidelity'.

 

As I mentioned, if you're ever able to listen to live music that does help. After that, I think it's really a matter of what you want to be able to afford.

 

Look at it this way - we realize that a 10% increase in sonic resolution and spatial imaging is going to cost 100% more than what we've already invested.

 

In other words, at some point you have to tell yourself that you're on a budget and maybe it's just better to buy more records.

 

It's not a defeatist outlook, more one of pragmatism.

 

A good analogy would be with the vehicles we drive. Would I rather have the latest model Ferrari instead of my FJ Cruiser with the studded snow tires? Hell yes! But I can't afford the Ferrari and even if I could it wouldn't make it up my driveway.

[sandyk]

I think SandyK (Alex) said it best in another thread without the direct quote but he suggested that enlisting emotions in a listening evaluation along with longer sessions was a more accurate evaluation than ABX or DBT.

It's quite possible to listen to 2 versions of the same track right through, perhaps through different equipment, and one sounds good, but the other subconsciously sets your toes tapping as well.

[esldude]

At the risk of offending Bill Scott, I would refer to the extensive experience of Barry Diament, which I believe contradicts your suggestion.

 

No it actually doesn't. Maybe Mayhem has slightly different ideas in mind.

 

I will explain myself a bit more on the TAS idea. It is surprisingly easy to put up a couple of good mics, record some good musicians in space and play that back to get something broadly similar. If you were there it might seem more than broadly similar (but you know where everyone was). Nevertheless true reproduction has some theoretical and practicial problems when you only have two channels. If you weren't there, and subjectively listen to a recording broadly similar or so dissimilar as to be unnatural is about all you can say. Even if well acquainted with natural musical instruments played live.

 

Yet I have read discussions of hall portrayal, space, real dynamics, details of forwardness or backwardness. You have no way of knowing which of those is right. Even with the finest experience of live music you can't actually determine that. I can tweek by about a decibel a couple octaves in EQ and push a singer forward or back in what you hear. Neither will be right and you have no way of knowing what the truth was. Your single ended subjective discernment is without anchor. And that is without even getting into the slim chances of getting an unmolested simple recording from commercial sources in the first place. Or how deft some recording professionals are at making a natural sounding effect you don't realize is an effect instead of real.

[Jud]

And how would we know the 'way' it was recorded?...........What's the reference?

 

If Audio and the audiophile hobby is of a subjective nature, all reference to anything is void, including 'Hi Fidelity'.

 

I think like most hobbies - cars, wine, dating - it's unavoidably subjective to some extent. Doesn't make purchasing a car because I like the color and the way it handles (without referring to skidpad tests) void.

 

I kinda just muddle through, working with a couple of things:

 

- I think I'm fairly familiar with the sounds of vocals, acoustic guitars, pianos. There are of course lots of varieties of these, and yep, I don't know how they were recorded, but I listen for what sounds natural to me, however unreliable that is.

 

- I see whether I get bored or irritated with the sound of a software player or piece of equipment over a period of time. The less "sound" of its own the player or equipment has - the more varied the sounds of what I'm listening to - the more accurate it is, allowing all the different recordings to sound like themselves. On the other hand, the more the player or equipment makes everything sound the same, the quicker I lose interest or get bothered. So boredom or irritation is a great indicator. (I hope that's reasonably clear; it's not always an easy idea to convey.)

[One and a half]

Listening to music is training our memories much the same as a diver learns how to double somersault, with a pike and twist of lemon. Repetition and learning from mistakes is the way to learn, the brain joins several separate smaller activities that it knows instinctively and records, and plays back what we want to do.

 

Imagine our memory is like an audio recorder on continuous record, taking in all sensory inputs, sight, sound, touch and so on. All that information is digestible to a certain degree, but there's a limit.

 

Sadly, our hard disk in the brain has limited memory capacity, so unimportant information is discarded and fragmented. for example boring threads on CA like the Sarcasm entry is out of my mind, since it offers no use for enjoyment or survival.

 

For musicians, their livelihood and offering their talents for others to appreciate, relies on audio memory of a high calibre and they are trained for that.

 

Same for audiophiles, long term listening is repetition and familiarity in one's own surroundings, and the sounds we hear are recorded deeply for instant recall if necessary and changes are readily apparent. So when something sounds wrong in the system it is wrong, for real.

[sandyk]

Same for audiophiles, long term listening is repetition and familiarity in one's own surroundings, and the sounds we hear are recorded deeply for instant recall if necessary and changes are readily apparent. So when something sounds wrong in the system it is wrong, for real.

 

+1