Absolutely true. Its how we understand speech in a noisy restaurant, and how we extract signal from noise in many other situations. Reminds me some of Kalman filtering. Also reminds me of compressed sensing (engineering).
The fact that there is a learned model to match a noisy signal against can go either way in terms of helping/hurting reception accuracy, but on average its evolved to improve accuracy in noisy conditions if correct prior learning has occured, IMO.
Yah I know, but there are people trained to ‘see’ things most do not......so who’s to say one can’t train to hear things most cannot?
There is ZERO research showing such a hypothesis to be true. However, there is good research evidence to show that 'expectation bias' is real and is an affliction of experimenters such as those who perform hearing research on test subjects.
I suppose like anything else it depends on how well trained one is. Or naturally inclined for that matter.
It's never happen to you, really? Have you never read something and then reread it and your initial reading was wrong and you realise it's because you read what you expected to read, for example?
Reading is somewhat different from listening in some ways, IMHO. But, I see what you are getting at.
Regarding listening, we have been through this over and over again, I absolutely do not rely on one listen by me and form a conclusion from that. Jam uses a protocol of repeated listening over three days. I sometimes ask other people to take a fresh listen in order to get a 2nd or even 3rd opinion, and I also find that listening repeatedly over multiple days is required for best reliability.
Let's take your reading example, have you ever read something carefully several times over in one day to make sure you understand exactly what it means, then checked again multiple times the next day, then asked other people what it means according to their reading of it, etc.
I don't come here after one quick listen and tell you I know how something sounds, and I sure didn't and could not ever sort opamps by distortion double blind in such an undisciplined and haphazard way as you suggest to perform reading.
Last edited:
The mistake is not in understanding but in reading the actual words as they appear on the page. Perhaps I missed you point.
My reply was by way of a comparison to listening to a conversation in a noisy room.
The text equivalent of listening in the presence of noise would be something like reading those distorted looking letters and numbers that people sometimes have to type in to prove they are not a bot.
Those distorted symbols are text buried in various kinds of visual noise. The brain can pick them out more accurately than the typical AI can. Its because the brain has learned more sophisticated pattern recognition.
In the case of listening, hearing distortion and other artifacts in audio reproduction is a form a of aural pattern recognition IMO, as I have stated before.
Interesting, but my point is that an expectation can happen, maybe depending on what has preceded or the environment etc, maybe it's more to do with language and not a good analogy for music.
To be even more fair, this is an unfair comment to make. Given your track record, if someone else wrote the converse of what you did above (no one will believe a DBT failing to refute the null hypothesis if it contradicts their views), I would hope you would contradict them.
We run the gamut of ability to adapt to new evidence, no matter one's background.
So what is Jam's reference(s)? Does he visit live, none amplified music events? You have avoided answering that question but maybe you can share som light of the masters habits.
//
//
This discussion can go on forever without any progress.
Conclusion is simple: some hear things that others don’t.
So what’s the big deal, I couldn’t care less.
I’m just happy for both.
Hans
Conclusion is simple: some hear things that others don’t.
So what’s the big deal, I couldn’t care less.
I’m just happy for both.
Hans
Let's let Jam (everyone) speak for himself, everyone. It's problematic enough that we all live in our own heads, much less trying to project someone else's brain forward.
The past couple pages of this thread are an embarrassment of terrible logic and argumentation. (all parties) Wide enough to drive a Panamax boat through.
The past couple pages of this thread are an embarrassment of terrible logic and argumentation. (all parties) Wide enough to drive a Panamax boat through.
Indeed, there is NO baseline. The only way we have learned to communicate quantitatively is through measurements. Sure, if something is heard, you can potentially teach another to hear it...but WHAT IS IT??? Without that knowledge, you cannot design for it, analyze it, or remove it if it is objectionable.
Designing audio by "I think I hear something" is akin to the era in astronomy before telescopes. All was conjecture, and almost all was wrong.
Circuit details and their strengths and weaknesses are what drive the state of the art forward. The commercial designers here know this, as is obvious by their designs. That doesn't stop them or ad agencies from embellishing in order to drive sales, but that part has nothing to do with performance.
Cheers,
Howie
I must have missed that - was that test set reported on DYIA? Or somewhere else it can be read?
Absolutely... I prefer this, or I prefer this because xxx, are fine and good!
I prefer this and I'm right and you are wrong, isn't ok.