Robert Harley's Wonderful Defense of our Subjective Hobby

[Ron Resnick]

In the January 2024 issue of the absolute sound Robert Harley writes a wonderful editorial about the importance to our subjective hobby of listening for oneself:

 

[GSOphile]

Kudos to Mr. Harley!

 

[henrich3]

Objectivists are skeptical of sighted listening tests because we know from experience that expectation bias can lead us to draw false conclusions about the device under test. Many may prefer the sound of a beautiful high-end speaker cable over a generic of equivalent AWG when we can see what we're listening to, for example. Listening test results can be quite different when we're denied that knowledge via ABX or double-blind testing however...

 

[Tangram]

I don’t see how anyone could take offence to Mr. Harley’s advice. Some people will hear a difference, like the difference, and deem it a worthwhile upgrade, some with hear a difference but not like the difference, and some will hear no difference at all. It’s up to the individual to be mindful of his biases - not the “bias police” - since he’s the one whose cash is at stake.

 

[Mike Lavigne]

over time, and experience, some of us learn to trust our ears and our personal process. and we never stop learning and enjoying.

others of us decide from the get-go that we can't. and insist on talking about that.

and never the twain shall meet.

 

[henrich3]

In *sighted* listening tests, some people may think that they hear a difference, like the difference, and deem it a worthwhile upgrade.

With properly controlled blind listening tests, the differences that we hear are more trustworthy.

I'm all for listening tests. We just need to do them blinded if we want to prevent expectation bias from influencing our conclusions. If you believe that device "A" sounds better than "B" 8 out of 10 times in blind listening tests, for example, then you can have confidence that the improved sound is real & not imagined. Expending this level of effort may be difficult for Joe Audiophile, but equipment reviews in magazines like TAS certainly should do this if they want greater credibility from their more skeptical readers.

https://seanolive.blogspot.com/2009/04/dishonesty-of-sighted-audio-product.html

https://nwavguy.blogspot.com/2011/05/subjective-vs-objective-debate.html

I read an anecdote years ago about testers who deceived listeners into believing that they were listening to high end speaker wires when in fact they were listening to generics and visa versa. I probably don't need to tell you which ones the audience preferred...

(Edited to insert links.)

 

[Al M.]

Thanks for posting, Ron.

Great defense. I am a scientist myself, but I am all for subjective evaluation as well.

Some, not all, engineers have a cocky attitude: they believe everything that needs to be measured can be measured, and we know exactly what to measure, all falsely in the name of "science". That's not how a scientist thinks.

A scientist knows that the more we know, the more we know that we don't know everything. If there is a discrepancy between what we are hearing and a measurement, it is not necessarily because the measurement is more "scientific". It could be that what we measure is simply flawed. Just like the first digital engineers did not know the incredible psychoacoustic impact of digital jitter (in the parts per billion) and proclaimed digital measures ("and sounds") perfect. A little humility is advised (see also Robert Harley's example about interface jitter in his essay). We don't know everything, and the digital engineers back then certainly didn't (fortunately we now know much more about jitter and how to manage it). Same thing holds for phenomena that we do still hear today but cannot quite measure yet, or if we can, not always do.

Fortunately there are enough engineers that are actually science-minded. Not all are. The cocky ones are on the ASR forum and the like, and spew their unscientific nonsense in the name of "science".

There is enough evidence that it s still best to trust our ears as the final arbiter.

Yes, measurements are useful, and no professional audio designer can live without them. But they are not always the Last Word.

 

[Gregadd]

We can see the pitfalls of personal preference. There is an objective truth. Even if, we don't know what is. We must however trust our own ears because we have no choice. It matters what we hear, not what a machine hears. It is the listener who must be convinced. However, if we rely solely on personal preference, in the end we wind up hiring the woman with the qqeong system.

 

[andromedaaudio]

I’m confused. Are you saying this WASN’T the ideal outcome?

What was the job ???

 

[jtcfdog]

Expectation bias is a factor sometimes I'm sure,as henrich3 mentioned.But there have been multiple times for me when trying a new component where I was hoping for an improvement that didn't come to fruition or actually sounded bad.Inferring that all audiophiles are gullible is rather biased.

 

[henrich3]

I see expectation bias as a subconscious thought process that allows us to pre-judge the merits of something novel. That's not the same as gullibility. It's just worth being aware that when most of us see a beautiful & expensive piece of audio gear we are naturally predisposed to believe that it will perform better than a less expensive pedestrian-looking device. That bias can trick us into believing that we're hearing audible improvements that don't actually exist. Of course that beautiful & expensive piece of audio gear may well provide real audible improvements, but we shouldn't pre-judge an outcome, even subconsciously, going into a test.

 

[jtcfdog]

Agreed.I'm doing some pre judging myself by assuming the majority of audiophiles won't be tricked that easily. I wish all of my modest components had the appearance of McIntosh,though their house sound is not for me

 

[JackD201]

I see expectation bias as a subconscious thought process that allows us to pre-judge the merits of something novel. That's not the same as gullibility. It's just worth being aware that when most of us see a beautiful & expensive piece of audio gear we are naturally predisposed to believe that it will perform better than a less expensive pedestrian-looking device. That bias can trick us into believing that we're hearing audible improvements that don't actually exist. Of course that beautiful & expensive piece of audio gear may well provide real audible improvements, but we shouldn't pre-judge an outcome, even subconsciously, going into a test.

Expectation bias is real. That said I don't see where it comes in if the device in question has been bought, paid for and lived with.

Evaluation under blind conditions still only yield snap impressions. I defend this by saying it only gives you performance based on a singular scenario by virtue of the controls. That scenario may or may not be representative of the variances of different uses and applications.

The bottom line for me is this. Do a blind test if you like. If you can't tell the difference go for whatever you like for the reason you like be it cheaper, better looking, whatever. There are no guns being pointed at anybody's heads and I personally don't need strangers coming around trying to save me from my self like a karen admonishing me for eating fries. Well, at least that actually helps me. LOL

 

[Al M.]

Objectivists are skeptical of sighted listening tests because we know from experience that expectation bias can lead us to draw false conclusions about the device under test. Many may prefer the sound of a beautiful high-end speaker cable over a generic of equivalent AWG when we can see what we're listening to, for example. Listening test results can be quite different when we're denied that knowledge via ABX or double-blind testing however...

Objectivists engage in pseudo-science. They only see numbers and measurements, and discount basic human psychology. Yes, expectation bias is part of that psychology, but so is stress or confusion when asked to perform under blind conditions. Blind test are suspect by virtue of that alone. Blind tests being "scientific"? Don't make me laugh.

There is a fine line between objective thinking and simplistic thinking. Objectivists too often cross that line.

Having said that, I have found no difference with an expensive power cord when listening for differences with stock power cord on my own under sighted conditions (a matter of system resolution back then, years ago,, and of context in which the power cord was used). That did involve repeated going back and forth on the same track at times, to make sure I did not fool myself when I heard differences (I always did).

On the other hand, I have easily and reliably identified in a blind test a superior audiophile power cord on mono block power amps in another system upon repeated choice on diverse music material (3 or 4 choices total, with random choice of which is A and which is B, and without disclosure which was A and which was B until the end). It did stress me out (my own power cord was on the line in that test), but I did it, flawlessly.

 

[henrich3]

I responded to this thread because I found Robert Harley's view of personal sighted listening experiences being the gold standard for evaluating audio gear to be inadequate. I brought up blind testing as a way audio equipment reviewers can control for the problem of expectation bias. If folks here aren't concerned about the issue, that's fine. I'm not looking for a fight. Just tossing in my two cents...

 

[cjfrbw]

I hear 'differences' all the time because I am a crazy audiophile and WANT to hear them. I will not recite the flagrant embarrassments chapter and verse that I have in the past immersed myself in. Too lengthy a list and too humiliating (not to mention expensive).
Sometimes I think the audio nostrums are immolative therapies for the insane. Where there is a market, there is somebody willing to make and sell the product.

I can state with pride, however, that I never fell for the quantum dot thing (preens self).

 

[Al M.]

I can state with pride, however, that I never fell for the quantum dot thing (preens self).

No? You should!

Beyond Just Technology: SuperUltraHyperTechnology

PYST cables are made from only the finest 6-nines Unobtanium™ alloy, molecularly assembled in our Alternate Universe™ reality-distortion tesseract field , using a secret geometry reverse-engineered from crashed UFOs, painstakingly smuggled out of Area 51 by deep-cover operatives. Performance is further enhanced by the use of a QuantConnect™ quantum-entangled pair of transmission interfaces, held at absolute zero by our exclusive Stasis Field™ technology. The cables are then wrapped in NanoAeroCap™, a nanotechnology-enabled aerogel anti-capacitance insulation system, featuring Fractal Interleaved Geometry™ to create negative inductance for maximum audio transmission quality.

Or, er, well . . . again, no. These are nice, high-quality cables, with solid, reliable connectors. That’s it. Hope you like them!

***

From the Schiit website, about their $ 20 Pyst cables.

 

[cjfrbw]

No? You should!

Wow! I'm sold.

 

[Al M.]

Wow! I'm sold.

Hehe, I knew you would be.

 

[mtemur]

I read Robert Harley’s writing on January 2024 issue and I can say I completely agree with him on this. Grounding makes a huge effect on sound that’s for sure, if you can’t hear a difference it’s because either your gear or ears are not good enough. Some prefer grounded, some prefer floated that’s another subject.

Additionally what kind of wood you placed under grounding cable also makes a difference. Simply everything makes a difference.