Post research here that casts doubt on ASR objectivism

[Andrew s]

Yes

 

[Frgirard]

Doesn't a theory still require a causal link, though? I'm not being snide, I'm actually really curious now.

Also, unrelated but we probably shouldn't be getting too down on the social sciences in this forum. Psychoacoustics happens to be one.

All in the audio folia is a good object of study for sociology, psychology see psychiatry.

 

[SubOjectivist]

Hello,

I don't know if this article has been posted here before, but just in case it hasn't:
https://www.researchgate.net/public...mulation_of_frequency_responses_of_headphones

I just had a (very) quick look, did not analyse its methodology since some (and most of those who replied to me) of you seemingly couldn't care less, so why bother?

Anyway, I think it could have some implications for both:
- S. Olive's study and hence the Harman circumaural headphone target curves, given that the different tested FR curves of various headphones were simply emulated,
- EQ in general, since apparently that could produce audible non-linear distortion, especially at levels > 80 dBA. Buth one study did use C weighting, while the other used A weighting.

What are your opinions on this matter?

 

[Wes]

Not sure I'd put Psychoacoustics into the social sciences category. Similarly, some of psychology is experimental - "scientific" but some is not. It's a big field.

Your definition requiring a causal mechanism would exclude things like comparative planetology, and Darwin's Origin theory. He did posit selection as the cause of species evolution, but had no knowledge of genetics and its particulate nature. So, it lacked a mechanism until Mendel's work was re-discovered. Or you could argue until DNA was discovered as the chemical form of the gene.

We always want a mechanistic explanation. Sometimes we have to wait for toxicologists to explain epidemiological theories in a mechanistic manner (3rd example).

Scientists do lots of things that are not within the hypethetico-deductive bailiwick. One reason why the Supreme Court threw up its hands when trying to determine what science "is" (for an evidentiary scale to allow in expert testimony).

And anecdotal evidence is important. We take photos of distant galaxies, send probes to other planets, and those initial forays are not much different than the Voyages of Discovery (Darwin was supposed to be a companion who could converse with the captain, and was not hired as a scientist), and Jefferson sent a bunch of English and French speaking white guys, a black guy, and an Indian woman to discover and collect specimens in the western half of the US.

 

[dc655321]

Also, unrelated but we probably shouldn't be getting too down on the social sciences in this forum. Psychoacoustics happens to be one.

The question wasn't intended as a dig. That was just the first area that came to mind.

There may be cultural/behavioural aspects of psychoacoustics, I don't know, but I do know it is primarily concerned with biophysics and perception.

 

[dc655321]

And anecdotal evidence is important. We take photos of distant galaxies, send probes to other planets, and those initial forays are not much different than the Voyages of Discovery

You're describing literal fact-finding/measurements with those astronomical examples. That's antithetical to anecdotal.

 

[Wes]

You're using a lay defn of anecdotal, not the one commonly used by scientists.

 

[rdenney]

I’ve been thinking about the causal mechanism thing. I have to admit that some of the most important “theories” I’ve worked on have not been based on a specific physical theory, but rather an attempt to find an elegant transfer function between known data and important effects. In other words, a model.

But I have drawn the line at models that specifically violated physical reality, such as (an example that comes to mind) distance to a fractional power. Those empirical models just can’t be considering the correct variables.

So, correlation can be rejected if it violates physical principles, it seems to me, even if the correlation is good.

What that has to do with audio—yeah, not much.

Rick “thinking of theories as models in the process of validation—maybe” Denney

 

[David Harper]

In my forty years of audiophilia IMHO it mostly boils down to this; "high-end" audiophiles assume a correlation between sound quality and price.
An amp that costs 10K must (and as a result does) sound "better" than one that costs 5K. And one that costs 50K sounds far better than either.
This nonsense is the result of decades of Stereophile and TAS influence. The best sounding component is the one most outrageously priced.
A quick study of the current turntable offerings from VPI along with the preposterous reviews that these get in the audiophile press is all that's required to know that this is true. People are, of course, free to spend (or perhaps more accurately waste) their disposable income on whatever they want. But it is, for the more seriously minded of us, a minor annoyance that so much of this kind of thing is evident and goes unchallenged in much of the audiophile press. I mean, if you're going to blow a small fortune on a toy at least let it be a toy that actually has an intrinsic value somewhat commensurate with it's price.

 

[audio2design]

The people I know in psychoacoustics are most definitely NOT in social sciences. They are looking at biological mechanisms, neurology, learning mechanisms, etc. That the results is not yes/no does not make it a social science.

 

[BrEpBrEpBrEpBrEp]

The people I know in psychoacoustics are most definitely NOT in social sciences. They are looking at biological mechanisms, neurology, learning mechanisms, etc. That the results is not yes/no does not make it a social science.

Psychology is a social science.

 

[dc655321]

You're using a lay defn of anecdotal, not the one commonly used by scientists.

I thought "anecdotal" here was a casual or experiential observation - no quantification or units associated with the observation.

 

[Wes]

could be casual - usually means not part of a systematized scheme

e.g. Darwin visits the Galapagos; AR Wallace in the Malay archipelago; the UC Berkeley training of how to keep a field notebook (into the 1950s); initial planetary probes are "informed by"* telescopic observations so are a middle ground

Contrawise, if observations are carried out in a statistical design in different ecosystem types, what have ya got? It's not experimental; it could be a test of a hypothesis with "natural" "controls" tho not manipulations, but it is arguably not anecdotal. Comparative techniques have been used in paleontology, economics, ecology, anatomy, physiology, anthropology, and on and one thru the harder sciences and the softer, squishy ones.

* I have deftly incorporated a liberal arts phrase herein

 

[Roland]

If the null hypothesis is a changed to: “There can be audible differences between competently designed and adequately powered electronics (amps, DACs, streamers), and those that measure the same can sound different“. Is there evidence to the contrary?

 

[BluesDaddy]

If the null hypothesis is a changed to: “There can be audible differences between competently designed and adequately powered electronics (amps, DACs, streamers), and those that measure the same can sound different“. Is there evidence to the contrary?

Yes, that in numerous double blind listening tests, people have NOT been able to discern an audible difference to a statistically significant degree.

 

[rdenney]

If the null hypothesis is a changed to: “There can be audible differences between competently designed and adequately powered electronics (amps, DACs, streamers), and those that measure the same can sound different“. Is there evidence to the contrary?

The few examples that turn up and that I’ve reviewed are shown to either 1.) reveal purposely distorted or colored sound, or 2.) expose that they were not as adequately powered as originally claimed.

But those are certainly measurable effects.

And they involved amps into speakers, and speakers do sometimes present a challenging enough load to cause an amp driven nominally to become non-linear at some level.

I’ve never seen controlled testing that could reliably identify different DACs that measure to be practically transparent.

Rick “lossy compression schemes are another matter” Denney

 

[Wes]

I agree that if any effort is expended on electronics, that effort should be focused on power amps. Distortion into difficult loads, clipping/soft-clipping, crossover distortion, and I'm not sure what else to add to the "investigation list."

 

[SubOjectivist]

Hello,

I don't know if this article has been posted here before, but just in case it hasn't:
https://www.researchgate.net/public...mulation_of_frequency_responses_of_headphones

Anyway, I think it could have some implications for both:
- S. Olive's study and hence the Harman circumaural headphone target curves, given that the different tested FR curves of various headphones were simply emulated,
- EQ in general, since apparently that could produce audible non-linear distortion, especially at levels > 80 dBA. Buth one study did use C weighting, while the other used A weighting.

What are your opinions on this matter?

I'm still a newbie here (but you already knew that, right?) , so if possible, I'd like to have some expert opinions on this study, anyone?
It's a psychoacoustics study, so I had hoped it'd interest at least some members here.
@preload maybe?

 

[BrEpBrEpBrEpBrEp]

I'm still a newbie here (but you already knew that, right?) , so if possible, I'd like to have some expert opinions on this study, anyone?
It's a psychoacoustics study, so I had hoped it'd interest at least some members here.
@preload maybe?

It's already widely known on this forum that EQ can amplify distortion and make it adible/more audible. The headphone reviews on here touch on that.

 

[BrEpBrEpBrEpBrEp]

I like that you share different market research on this thread. But I don't understand why you lose your time. There are already some verified websites to find different predictions and research on the market in every domain. I am already informing myself on such websites for already two years. It is a quick way of gathering industrial information for every person in business. I am using marketresearchfuture.com, where I can always find the latest news about the market and its change. This is the first thing I read when I wake up in the morning.

what