Counter-Placebo Effect?

[GGroch]

OK I have had a look, and I suppose the answer it gives to my question is that we don't lose the perception of the placebo effect after being informed that we were in the placebo group and reporting an effect. It may well be modified, but can't be relied upon to disappear.

Right, I'm sure part of it is due to the fact that we are pre-conditioned to think pills have an effect, and that it is usually positive. If we are told that one does not have an effect, some of that pre-conditioning remains.

I think "Placebo Effect" in 2024 is not the best term for what is happening here. In common parlance it denotes "tricking" the brain, like an optical illusion. Perhaps an interesting parlor trick that impacts our lives infrequently.

In fact, it illustrates how our minds work as "Prediction Machines" all of the time. All of our senses are constructs of limited input from ears, eyes, noses, combining with massive amounts of pre-conditioning in our brains that interpret that input.

Oddball is right that it impacts all areas of our lives. Especially now in the age of social media. Pre-conditioning from an almost unlimited stream of mostly unreliable sources gives rise to "alternative facts" which help to explain how 1/2 the U.S. population thinks the other 1/2 is nuts.

For some, differences will be so marginal that they will be thrown in the placebo category, even though it will be theoretically on overall value perception (i.e. not just $ driven). For others, differences will be meaningful and might or might not be worth the price, depending on the $ perception/budget.

If I understand you correctly, I do not agree that placebo effects are generally "marginal". Listeners often report huge differences in clarity/detail, when comparing power cords where there is no actual performance impact. As the article I reference above states, the impact of placebos is often directly related to the price of the pill (or cord), the more you spend the more it improves your condition. Whether you are getting what you pay for is more directly tied to your finances and your suggestibility than any weakness in placebo effect.

 

[kemmler3D]

*anecdote:

There have been a handful of times when I have heard a noticeable (but smaller than expected) difference while turning a filter / PEQ knob while the filter was turned off / defeated. When I realized it was defeated, the heard difference stopped.

In those cases, finding out it was placebo eliminated the effect.

 

[Oddball]

Right, I'm sure part of it is due to the fact that we are pre-conditioned to think pills have an effect, and that it is usually positive. If we are told that one does not have an effect, some of that pre-conditioning remains.

I think "Placebo Effect" in 2024 is not the best term for what is happening here. In common parlance it denotes "tricking" the brain, like an optical illusion. Perhaps an interesting parlor trick that impacts are lives infrequently.

In fact, it illustrates how our minds work as "Prediction Machines" all of the time. All of our senses are constructs of limited input from ears, eyes, noses, combining with massive amounts of pre-conditioning in our brains that interpret that input.

Oddball is right that it impacts all areas of our lives. Especially now in the age of social media. Pre-conditioning from an almost unlimited stream of mostly unreliable sources gives rise to "alternative facts" which help to explain how 1/2 the U.S. population thinks the other 1/2 is nuts.


If I understand you correctly, I do not agree that placebo effects are generally "marginal". Listeners often report huge differences in clarity/detail, when comparing power cords where there is no actual performance impact. As the article I reference above states, the impact of placebos is often directly related to the price of the pill (or cord), the more you spend the more it improves your condition. Whether you are getting what you pay for is more directly tied to your finances and your suggestibility than any weakness in placebo effect.

I am distinguishing placebo from "marginal difference" that a fair amount of people sometimes (often) throw in the same pot of unmeaningful differences where it ends with full placebo. This is digital world as you noted, so there are only 2 pots. At large, analogue thinking at the end requires down-sampling to digital. People on this forum are not representative of typical population and know how to split a strain of hair in 10 or 100 - which is not necessarily how to world at large perceives these issues. For most a strain of hair, or 10 of them are marginal.

As to the power cords - it either works or it does not. I recently had a bunch of issues with cheaper ($5 but on its face well made) 10A power cords and my old Rotel 1070 amp (2x130W into 8ohm). Replaced with a more expensive 15A rated power cord and since then never tripped the circuit switch. Yeah amp obviously developed issues and attitude, like all of us do with age, but at least to me it was simpler to give it better power cord than deal with repairs and outage. On its face, can't tell that the new and more expensive power cord has a better grip on the amp power socket - but for whatever reason it works.

 

[Justdafactsmaam]

The placebo effect is well understood and, I hope, widely accepted.

And so is how it applies to the audio hobby, e.g. price placebo, brand placebo, mods-placebo, highres-placebo, etc

BUT

Has anyone seen studies into whether there exists a counter-placebo effect? (my term)

That is, when we are told that what we have been perceiving as different or better is only a placebo, do we lose the perception of the placebo effect?

Or do our biases continue to influence our perceptions?

Anecdotes are welcome, but please preface them with *Anecdote.

I am really interested in any studies, or well-formed demonstrations.

Cheers

I think it is important to differentiate between a placebo effect and the phenomenon that is often referred to as a bias effect in audio.

The placebo effect relies completely on belief. If the belief goes away so does the effect. If the belief is never there the effect is never there.

What is often referred to as “bias effects” in audio is a function of limited processing power of our brain to take in and absorb everything we hear. So we have to steer our focus. In steering our focus we filter what we take in from the sound we hear. Our steered focus can be consciously steered or unconsciously steered. And this is where bias effects can influence things. But our focus is always steered because of the physiological reality that we can’t process everything we hear.

So unlike the placebo effect there is no getting around or getting over bias effects or as we should call it, steered focus effects.

There certainly is overlap. But there is also a substantial difference

 

[Justdafactsmaam]

Depends on the placebo.

If a test is trying to figure out what race someone is based on the sound of their voice. The placebo instantly shatters the moment the revelation is made.

But if you're trying to counter the McGurk Effect, it won't happen. The mind will continue to lie to you based on what is visually occuring, over what is audibly being uttered.

The McGurk effect isn’t a placebo. It’s a biological priority system that evolved because it works. When two senses are giving conflicting data to our brain our brain automatically prioritizes the sense that gives us better, more reliable data.

Placebo is very different

 

[Newman]

What is often referred to as “bias effects” in audio is a function of limited processing power of our brain to take in and absorb everything we hear. So we have to steer our focus. In steering our focus we filter what we take in from the sound we hear. Our steered focus can be consciously steered or unconsciously steered. And this is where bias effects can influence things. But our focus is always steered because of the physiological reality that we can’t process everything we hear.

So unlike the placebo effect there is no getting around or getting over bias effects or as we should call it, steered focus effects.

I think the one specific bias effect you have described is just one of many cognitive bias effects, not a full description of the scope of bias effects in audio.

 

[Newman]

The placebo effect relies completely on belief. If the belief goes away so does the effect. If the belief is never there the effect is never there.

Sure, but the specific question I raised in post #1 (“That is, when we are told that what we have been perceiving as different or better is only a placebo, do we lose the perception of the placebo effect? Or do our biases continue to influence our perceptions?”) is not describing a situation where the belief ‘goes away’.

Example: sighted listening of cables where an audible difference is perceived but does not exist. We can tell such a person that there is no actual audible difference, but they might not believe us, so their belief doesn’t ‘go away’, so the perception endures.

 

[Justdafactsmaam]

Sure, but the specific question I raised in post #1 (“That is, when we are told that what we have been perceiving as different or better is only a placebo, do we lose the perception of the placebo effect? Or do our biases continue to influence our perceptions?”) is not describing a situation where the belief ‘goes away’.

Example: sighted listening of cables where an audible difference is perceived but does not exist. We can tell such a person that there is no actual audible difference, but they might not believe us, so their belief doesn’t ‘go away’, so the perception endures.

Yes. The difference with audio is in a non time synchronized quick switching test it might seem like you are hearing differences even if you don’t expect a difference. That’s because of steered focus. And that is the Achilles heel of audio. When you think you are hearing real and substantial differences when you didn’t expect to *that* makes for a very convincing personal experience.

 

[GGroch]

.....That’s because of steered focus. And that is the Achilles heel of audio. When you think you are hearing real and substantial differences when you didn’t expect to *that* makes for a very convincing personal experience.

In searching ASR for "steered focus" it appears you are the only member to have used the term, which would be very odd if it is "the Achilles heel of audio".

Certainly the brain's selective filtering can lead us to hear, see, or taste differences when none exist, and to ignore differences when they do exist.
This seems to impact all senses. Witnesses of an armed robbery are statistically far more accurate in describing the gun used than they are at describing the perpetrator's height because their mind focuses on the gun. As Newman says, this is one of many roads to cognitive bias.

But it does not describe why audio snake oil exists. Snake oil purveyors precondition customers with detailed descriptions of the supposed technical superiority of their products and then describe the differences they should listen for in advance of a demo.

The steering is done by the salesman. Customers are then usually surprised by the huge differences they hear, Their unconscious expectation bias was established by the salesman. It works because our brains are prediction machines not arbiters of truth. This bias is eliminated in a blind test, and the huge differences they hear disappear. The selective filtering bias you describe would not disappear in a blind test I think.

Establishing bias was how I sold audio gear and trained others to sell it. Predict, Prove, Confirm, Ask for the sale. To be fair, I thought I was educating customers at the time.

 

[Justdafactsmaam]

In searching ASR for "steered focus" it appears you are the only member to have used the term, which would be very odd if it is "the Achilles heel of audio".

That does not reflect very well on ASR.

Certainly the brain's selective filtering can lead us to hear, see, or taste differences when none exist, and to ignore differences when they do exist.
This seems to impact all senses. Witnesses of an armed robbery are statistically far more accurate in describing the gun used than they are at describing the perpetrator's height because their mind focuses on the gun. As Newman says, this is one of many roads to cognitive bias.

Yup. We are not human tape recorders

But it does not describe why audio snake oil exists. Snake oil purveyors precondition customers with detailed descriptions of the supposed technical superiority of their products and then describe the differences they should listen for in advance of a demo.

That’s called steered focus. The thing is we steer our focus one way or another. Either by expectation or by other influences. Even randomly. But it’s always steered. We can’t process everything we hear.

The steering is done by the salesman. Customers are then usually surprised by the huge differences they hear, Their unconscious expectation bias was established by the salesman. It works because our brains are prediction machines not arbiters of truth. This bias is eliminated in a blind test, and the huge differences they hear disappear. The selective filtering bias you describe would not disappear in a blind test I think.

A blind test only works if it is time synchronized and quick switching. Watch the video. JJ explains it really well.

Establishing bias was how I sold audio gear and trained others to sell it. Predict, Prove, Confirm, Ask for the sale. To be fair, I thought I was educating customers at the time.

That’s how powerful the experience is. Because it really does change how you perceive the sound.

 

[Madlop26]

“People can still get a placebo response, even though they know they are on a placebo,”

Yeah, that's what I feared.

So, even if we are completely satisfied that expensive power cable X can't possibly sound better, we still might hear it as better.

In which case, there is merit in buying it. Yes?

merit? hardly, a discovered placebo effect greatly reduced its power in the subject, but if still some remains, it is kind stupid to spend money on it. If your brain can be so easily subject to cognitive bias you can find cheap methods to get your placebo effect. I heard spreading moisturized holly water in your listening room increased soundstage by 50%, no kidding.

 

[DannyBoyNYC]

Interesting topic and not surprised with the findings that people can believe in anything even when pointed out the truth. Unfortunately happens more broadly than audio and in more fundamental areas of our life. It also gets worse, when something objectively worse can be perceived as better.

"The mind is its own place and in itself, can make a Heaven of Hell, a Hell of Heaven." - John Milton, Paradise Lost
Probably much more broad and fundamental than we can image if we confine our observations to audio. Interesting topic indeed!

 

[SIY]

A blind test only works if it is time synchronized and quick switching.

That's only true for particular test formats. A sorting test, for example, works perfectly well for suitable phenomena.

 

[Justdafactsmaam]

That's only true for particular test formats. A sorting test, for example, works perfectly well for suitable phenomena.

I was speaking strictly about audio and discerning differences

 

[GGroch]

That’s called steered focus. The thing is we steer our focus one way or another. Either by expectation or by other influences. Even randomly. But it’s always steered. We can’t process everything we hear.

I'm glad you posted JJ Johnston's presentation. It is as great an explanation of how signals from our ears/eyes/nose etc. reach our consciousness I have seen. I have posted this same video elsewhere on ASR.

The presentation is 11 years old it is all still valid and important. There is a lot of filtering done and our consciousness interprets from limited information.
JJ includes the attached graphic (I have added the red arrow):

JJ does not address here much of how "Cognitive-Level Understanding" works, and I think that is where biases and the mind as a predictive machine come in. This describes how the brain prioritizes its limited information and adds a lot to the discussion.

Most research on this is new, within the last 5 years. See YouTube Links on the Predictive Brain Here. It provides a better understanding of how the brain actually understands things. How it interprets the blend of current sensations with previous experience.

To summarize it, even if the road from our ears to our brains was perfect, even if we received all of the information unfiltered, we would still get things wrong because our brains are not designed to arbitrate truth.
Our brains are prediction engines designed to let us act quickly and win arguments. Bias is fundamental to how they work. This understanding better explains how Placebos work making it easier to understand why they are not a separate thing, but a part of cognition as a whole.

So... while understanding of how filtering works is critical, it is no longer seen as the only or even most important reason audio errors or any other cognitive errors exist.

 

[SIY]

I was speaking strictly about audio and discerning differences

So was I.

 

[Justdafactsmaam]

So was I.

If a dbt to discern audible differences isn’t synchronized with quick switching it’s a very flawed test. Level matching (if possible) time synchronization and quick switching are every bit as important as the bias controls.

 

[SIY]

If a dbt to discern audible differences isn’t synchronized with quick switching it’s a very flawed test. Level matching (if possible) time synchronization and quick switching are every bit as important as the bias controls.

Again, it depends on the experiment, especially the specific question to be answered. I gave several examples in my Linear Audio article.

 

[Justdafactsmaam]

Again, it depends on the experiment,

The only one’s I know of for determining audible differences are ABX and ABChr. Both of which depend on being level matched (if possible) time synchronized and quick switching

especially the specific question to be answered.

Is there an audible difference?

I gave several examples in my Linear Audio article.

Where can I find that article?

 

[SIY]

Is there an audible difference?

That's not a specific enough question to design an experiment. Between what and what? By whom? Under what conditions? These are all VERY important for doing good experiments.

The only one’s I know of for determining audible differences are ABX and ABChr.

There are more.

Indeed, rapid switching and good synchronization are necessary for the ABX format, but not sufficient.