The fallacy of Expectation Bias

[JackD201]

Well, judicious tweaking would have involved opening the speakers up and installing some serious internal cross bracing. Since I barely got through shop and love my digits, doing so would not be what I consider "judicious". That speaker is a case of woofers over powering their own cabinets! LOL

Funny thing is, the budget F series Klipsch speakers aren't shrill at all. In fact their blunted to a degree. They put a cover over the diaphragm of the tweeter that actually tamed the metal domes inside. Just for a semblance of being on topic, I was expecting them to be bright, especially when mated to a Denon receiver on the bottom of the Denon totem pole. The midrange when doing spoken word is very, very clear. The frequency extremes however...............

 

[fas42]

A nice agreement at least on one point: it was the F-2's I was listening to, the F-3's weren't as balanced in their "raw" form. And, yes, with a bottom rung Denon would be a good match. Frequency extremes are what it's all about, this is where the tweaking has to climb on board: to put things in context, some of the most vicious assaults on my eardrums have come from Wilson speakers - I pick up the shreds of skin from the floor at the end of the session. So, to also be "topical", I expect this brand to give me a hard time unless "tamed" ...

Frank

 

[Orb]

Coming back to expectation bias.
It should be noted what expectation bias is, and incredibly important just how it affects us in scientific terms.
Unfortunately too many aspects of bias and behaviour are put down to this very specific bias, that is tied medically into placebo affect, and importantly conditioning/false induced pain/etc tests.

The 1st problem for us as general consumers is that it is very difficult to overcome expectation bias as we condition ourselves to enjoyment-gratification and also pain (which has been tested to see how we behave to conditioning and the expectation of say a mild shock, which then seems much worse or a reaction to the pain button pushed even if no shock is applied).
The conditioning aspect of expectation bias can be seen in drug addicts as they feel an initial buzz-sensation before actual intake (this has been monitored and involves a chemical release), so the expectation causes a chemical reaction that causes enjoyment and enhanced reactions.

This same behaviour-reaction is seen in medical trials and why we have the placebo affect; modern research is showing that again we have chemical reaction in the body and not just in the brain, which again affects subtly mood and also lessens pain (which is what they were measuring and monitoring).
This is one reason why we have DBT, the patient must be involved in a drugs trial without the conditioning and chemical reaction caused by expectation of receiving medication as this skews perceived results due to the affect of the body's own chemical release.

So expectation bias is not that we select the most expensive product from a cognitive bias perspective (different bias), but that we have conditioned ourselves to expect greater gratification from more expensive products (hence the conditioning) that then causes chemical release that improves our mood and enhances our enjoyment.
So from a perception perspective the consumer is right that it sounds better to them, but this may mostly be down to the consumer equating enjoyment-pleasure to better, and this view may change when the conditioning-chemical reaction lessens.

The concept of the above paragraph has been also applied to a marketing study of wines, where they monitored mOFC (chemical release in the brain) and this affects enjoyment.
In said study of mOFC, a trend was noted between our expectation (gratification) of a wine's price and how much pleasure-enjoyment the glass provided.
The result was no different to what I explained above, with mOFC increasing just before drinking what they thought was a more expensive wine.

What I am getting at is this; expectation bias is a combination of conditioning and chemical release in the body/brain that subtly affects our mood and gratification (pleasure-enjoyment).
Hence, then why a more expensive product can seem better in the short term, or even repeated processes-cycle such as enjoyment of even selecting and cueing up an LP while having a drink.

Selecting a more expensive product involving a different but cognitive bias is not the same; it could as an example involve cognitive anchoring where we know some expensive products are good and associate-reference generically other expensive products to those, when in reality there is or should be no association.
Good example of this type anchoring; if I flip a coin 10 times and it comes out heads 8 times, many would place a high bet thinking it is more likely heads will result again, in reality it is 50/50.
Similar behaviour is seen in casinos with people playing roulette,etc.

I will not mention in this post about possibility and that biases can be overcome to some extent (interesting to consider if this applies equally to both cognitive and chemical-biological related biases and conditioning).

Cheers
Orb

 

[FrantzM]

Coming back to expectation bias.
It should be noted what expectation bias is, and incredibly important just how it affects us in scientific terms.
Unfortunately too many aspects of bias and behaviour are put down to this very specific bias, that is tied medically into placebo affect, and importantly conditioning/false induced pain/etc tests.

The 1st problem for us as general consumers is that it is very difficult to overcome expectation bias as we condition ourselves to enjoyment-gratification and also pain (which has been tested to see how we behave to conditioning and the expectation of say a mild shock, which then seems much worse or a reaction to the pain button pushed even if no shock is applied).
The conditioning aspect of expectation bias can be seen in drug addicts as they feel an initial buzz-sensation before actual intake (this has been monitored and involves a chemical release), so the expectation causes a chemical reaction that causes enjoyment and enhanced reactions.

This same behaviour-reaction is seen in medical trials and why we have the placebo affect; modern research is showing that again we have chemical reaction in the body and not just in the brain, which again affects subtly mood and also lessens pain (which is what they were measuring and monitoring).
This is one reason why we have DBT, the patient must be involved in a drugs trial without the conditioning and chemical reaction caused by expectation of receiving medication as this skews perceived results due to the affect of the body's own chemical release.

So expectation bias is not that we select the most expensive product from a cognitive bias perspective (different bias), but that we have conditioned ourselves to expect greater gratification from more expensive products (hence the conditioning) that then causes chemical release that improves our mood and enhances our enjoyment.
So from a perception perspective the consumer is right that it sounds better to them, but this may mostly be down to the consumer equating enjoyment-pleasure to better, and this view may change when the conditioning-chemical reaction lessens.

The concept of the above paragraph has been also applied to a marketing study of wines, where they monitored mOFC (chemical release in the brain) and this affects enjoyment.
In said study of mOFC, a trend was noted between our expectation (gratification) of a wine's price and how much pleasure-enjoyment the glass provided.
The result was no different to what I explained above, with mOFC increasing just before drinking what they thought was a more expensive wine.

What I am getting at is this; expectation bias is a combination of conditioning and chemical release in the body/brain that subtly affects our mood and gratification (pleasure-enjoyment).
Hence, then why a more expensive product can seem better in the short term, or even repeated processes-cycle such as enjoyment of even selecting and cueing up an LP while having a drink.

Selecting a more expensive product involving a different but cognitive bias is not the same; it could as an example involve cognitive anchoring where we know some expensive products are good and associate-reference generically other expensive products to those, when in reality there is or should be no association.
Good example of this type anchoring; if I flip a coin 10 times and it comes out heads 8 times, many would place a high bet thinking it is more likely heads will result again, in reality it is 50/50.
Similar behaviour is seen in casinos with people playing roulette,etc.

I will not mention in this post about possibility and that biases can be overcome to some extent (interesting to consider if this applies equally to both cognitive and chemical-biological related biases and conditioning).

Cheers
Orb

Stop making Sense ! Orb

Seriously .. Superb Post .. Your usual balanced and educative posts...

 

[Gregadd]

if I flip a coin 10 times and it comes out heads 8 times, many would place a high bet thinking it is more likely heads will result again, in reality it is 50/50.

I get the point. Personally having gotten 8 heads in a row, I would tend to bet the next would be tails. I digress. A knowledge of math teaches us that each coin flip is an indepedent event and the previous result has no effect on the current one. That would knock out any expectation bias. It proves my point that we can learn.

Of course relying on price alone will lead us astray. In designing a product we know that there are techniques and materials that can improve the product and that many are more expensive.

 

[microstrip]

Reading about expectation bias I found an interesting paper about Minimizing expectation bias in fire investigations. http://www.google.com/url?sa=t&rct=j&q=expectation%20bias%20&source=web&cd=6&ved=0CEsQFjAF&url=http%3A%2F%2Fwww.firescientist.com%2FDocuments%2FMinimizing%2520Expectation%2520Bias%2520in%2520Fire%2520Investigations%2C%2520ISFI%25202008.pdf&ei=x2QDT97dAcyD-waeppjNAQ&usg=AFQjCNECYo9ht2b4vcCSu1hdtZUflVTxKQ&cad=rja

The short introduction of the article about bias expectation in general is worth reading, but a particular sentence attracted my attention.

In the long-term, science is generally self-correcting. That is to say that the truth eventually emerges, or at least evolves, as our understanding of the universe improves. This built-in self-correction alone may suffice in many cases; however, for dealing with applications of science to shorter-term inquiries, the inherent self-correcting nature of science cannot be relied on exclusively. It needs to be supplemented and augmented by additional correction mechanisms and safeguards.

IMHO expectation bias is part of the audio game. No one can completely resist to it. The practical constraints associated with the techniques needed to overcame the listening biases using DBTs made them impossible to use in daily evaluation of high end systems. If we oversimplify them, reducing the time, number of tests and number of variables in the test, the tests risk to conduce to a false result.

This can indicate that the solution of our debates can be waiting unless a generally accepted sound reproduction theory scientifically explains all we want to know about sound reproduction, and postpone our buying decisions until we take them rationally, based on science.

But if we want to go faster, and buy something soon, we can thing about additional correction mechanisms and safeguards. Can you make suggestions and implementation methods for these practices, that can be used regularly by end users?

 

[fas42]

One thing I do when I listen to a new system for the first time is to not listen to it. There are screeds of advice and comments about how to "tune in", and "understand" what a certain system was set up to achieve. I couldn't care less about such matters: I almost instinctively ignore the system, treat it as a muzak machine and focus on talking to the people there and other matters. The sound of the system then filters in sideways into my consciousness, and that tells me a lot about the inherent quality of the setup: if it's running loud and the sound just flows over me that's a big thumbs up, but if, as typically is the case, the "noise" just irritates the bejeezus out of me, then the system has major problems.

Frank

 

[NorthStar]

I still remember when I broke up with my first girlfriend.
Then I found another girlfriend not too far after that, and I thought, while I was unwrapping her,
that she would be much better, with her two big and beautiful green eyes, and all that Jazz...

Well, it took me only a very short time to realize that my expectation was only vapor...
She would beat me up repeatedly!

That was over forty years ago.
Today I am much more down to earth, and I have acquired some wisdom to go with it...

:b

 

[amirm]

Bob, as you were telling the story with the green eyes and such, I thought the punch line would be that you were talking about Tree!!! Having had tree branches snap at my face after I cut part of them, I guess it is still possible you were talking about that and not real people .

Orb, superb insight there on expectation bias!

 

[treitz3]

There has got to be some exceptions to expectation bias, I would assume. I say this because I have purchased the expensive, listened to the insane expensive, auditioned the expensive and the same goes to the mid-grade and the cheap. While I'm admitting that their may be some sort of initial expectation bias at first, reality soon kicks in and things are what they are. From trucks to audio to sex to cooking or whatever it is I may encounter.

I have heard $120,000.00 MBL setups that were musically stomped by a cheap little Jolida integrated and a pair of Noel bookies. I have had [at the time] my dream amp in my rig that even before every amount of tweaking, tube rolling, cable swapping and what not....the expectation bias quickly turned into reality. Needless to say, it didn't last very long in my rig at all and I'm sure you could imagine the expectation bias on that one. Once again, things are what they are. Forgive me for the assumption but it is almost as if everyone scientifically, according to what I have read in this thread, has this so called expectation bias. Personally I disagree and quite honestly, we can agree to disagree. That's fine.

IMO, money does not mean anything in this hobby [though, it does help to achieve what it is you may be looking for]. So long as I realize this, their is no expectation bias. Another thing I am admittedly guilty of is knowing that what may sound good to one, may not sound good to another and that empirical evidence may or may not match your tastes in reproductive qualities. Add the aforementioned into the equation of things are what they are and expectation bias goes flying out the window.

So I guess what I'm saying is that expectation bias may affect some folks [general public maybe] but there has got to be exceptions to the rule because I can't think of many cases where expectation bias has been an influence on me. Things are what they are and I have no issues trusting my ears.

 

[fas42]

No worries, treitz3, I'm on your side. I think at times this "bias" thing is pulled out by people as an easy explanation for why other people's experiences don't match up with their own: we human organisms are complex beasties, so there are all sorts of factors that contribute to how people experience and process sensory input, and the ear/brain is as least as fancy and convoluted as any of the other senses.

We are all fooled at times into thinking something is better or worse than it actually is on reflection, and on further experiencing. But long term acquaintanceship with all manner of things in the end will win out in allowing one to derive a much truer assessment of the experience. Perhaps not the "absolute truth", which in nearly every case is a meaningless and unreachable ideal, but a hell of a lot better than anything that the tag of "expectation bias" can be attached to ...

Frank

 

[NorthStar]

Bob, as you were telling the story with the green eyes and such, I thought the punch line would be that you were talking about Tree!!! Having had tree branches snap at my face after I cut part of them, I guess it is still possible you were talking about that and not real people .

Orb, superb insight there on expectation bias!

See Amir; even you are bias towards a "Tree" when reading some of my posts!

By the way, do you know from what 'Posts' are made of?

 

[JackD201]

When I scanned your post and saw "unwrapping" and "her two big.." I definitely wasn't expecting green eyes!

 

[Orb]

I get the point. Personally having gotten 8 heads in a row, I would tend to bet the next would be tails. I digress. A knowledge of math teaches us that each coin flip is an indepedent event and the previous result has no effect on the current one. That would knock out any expectation bias. It proves my point that we can learn.

Of course relying on price alone will lead us astray. In designing a product we know that there are techniques and materials that can improve the product and that many are more expensive.

Yeah glad you see the point that even going with tails it is still an anchoring bias and the subtle difference between this and actual expectation bias and potentially with its medical placebo effect (something else used too generically IMO outside of my earlier explanation).
Indeed you are correct that it IS possible to overcome cognitive biases and there are some research showing some aspects of methodology to do so (posted on this subject in various other bias threads including papers).
The key is disciplined and open mind while using a form of self assessment-perception, so this enables defining and importantly using-following the framework while understanding bias mechanisms and behaviour.
More detail though can be found in other threads where I mention this.

What I am really unsure about though is overcoming chemical related biases triggered in the body and brain; when considering perception combined with enjoyment-preference this specific aspect is very interesting in how it affects our listening and listening behaviour, and of course related biases.
In theory reducing the stimulation-chemical response would help to mitigate such perception enhancements, and this could be done with understanding/mitigating the trigger-conditioning for a listener.
However that is just theory, challenge is for a person to be able to follow strict framework in same way of reducing cognitive biases (and even that is a real challenge for most).
Having an analytical mind unfortunately is not enough as biases can still be seen in operation even with those that have a scientific/mathematic/engineering background, very challenging.

That said, in audio some of the journalists I have the most respect for though are those with scientific research backgrounds combined with masters-phd usually either Chemistry or Physics, with some outside these traditional sciences such as Kal.
It is interesting studying the approach and writing from contributors such as John Atkinson, Paul Miller, Keith Howard.
All three share similar traits IMO when it comes to writing-discussions-forums, and all three have a very similar background, Kal also is very similar to these IMO (also with a comparable high scientific background).
It stems from what seems to me classical science education at masters-phd level combined with further research work, without this it seems some who are research scientists lack the same level of bias resistance.
However this is still oversimplifying why some seem better at overcoming biases as those I mention seem to be more of the exception.

Just my take on this anyway but sorry to digress a bit, and I appreciate there are more than those I named who have similar backgrounds or approaches.
Frantz, glad you enjoyed the post -at least someone reading them
Cheers
Orb

 

[Phelonious Ponk]

There has got to be some exceptions to expectation bias, I would assume. I say this because I have purchased the expensive, listened to the insane expensive, auditioned the expensive and the same goes to the mid-grade and the cheap. While I'm admitting that their may be some sort of initial expectation bias at first, reality soon kicks in and things are what they are. From trucks to audio to sex to cooking or whatever it is I may encounter.

I have heard $120,000.00 MBL setups that were musically stomped by a cheap little Jolida integrated and a pair of Noel bookies. I have had [at the time] my dream amp in my rig that even before every amount of tweaking, tube rolling, cable swapping and what not....the expectation bias quickly turned into reality. Needless to say, it didn't last very long in my rig at all and I'm sure you could imagine the expectation bias on that one. Once again, things are what they are. Forgive me for the assumption but it is almost as if everyone scientifically, according to what I have read in this thread, has this so called expectation bias. Personally I disagree and quite honestly, we can agree to disagree. That's fine.

IMO, money does not mean anything in this hobby [though, it does help to achieve what it is you may be looking for]. So long as I realize this, their is no expectation bias. Another thing I am admittedly guilty of is knowing that what may sound good to one, may not sound good to another and that empirical evidence may or may not match your tastes in reproductive qualities. Add the aforementioned into the equation of things are what they are and expectation bias goes flying out the window.

So I guess what I'm saying is that expectation bias may affect some folks [general public maybe] but there has got to be exceptions to the rule because I can't think of many cases where expectation bias has been an influence on me. Things are what they are and I have no issues trusting my ears.

Perhaps the expectations, and biases, are based on something other price. Perhaps they are changeable, situational. Perhaps no amount of cork sniffing, leg-watching, and color evaluation can live up to the expectations if what hits the pallet is vinegar. That doesn't except one from the rules, it just allows one to over-simplify them and continue to expect that they are immune.

Tim

 

[treitz3]

Perhaps the expectations, and biases, are based on something other price.

Hello, Tim. You are right. In the case of audio and myself, everything is based upon sound. Generally seeing how [if I am expected to have any sense of expectation bias] I have not heard said component before then what kind of bias should I expect? If there's no way to tell if something is going to sound good until one gets that component in said rig, then I can not understand where or what expectation bias can be expected.

With regards to immunity, does the expectation bias derived from science dictate that we all must have expectation bias?

 

[Phelonious Ponk]

Hello, Tim. You are right. In the case of audio and myself, everything is based upon sound. Generally seeing how [if I am expected to have any sense of expectation bias] I have not heard said component before then what kind of bias should I expect? If there's no way to tell if something is going to sound good until one gets that component in said rig, then I can not understand where or what expectation bias can be expected.

With regards to immunity, does the expectation bias derived from science dictate that we all must have expectation bias?

If "everything is based on sound" not on any information you've heard about the company, component, design, it's reported compatibility with X, etc, etc. If you bring no knowledge of the product to the first listening and are utterly uninfluenced by appearance, then you are not subject to expectation bias, because you have no expectations. If you have any of the knowledge of any of the above, and are still completely unaffected by expectations, then you, yourself, should be the subject of a scientific study because you are, indeed, an anomaly.

Tim

 

[Ron Party]

It amazes me how far some people are willing to go to discredit the scientific method in order to attempt to validate their opinions. Other than faith/religion, to the extent it purports to explain the physical world, is their any other aspect of the perceptual world where people go to great lengths to discredit the scientific method?

WRT expectation bias, it certainly can be the case that one can be immune from its effect in a comparison undertaking. OTOH, if one wants to be certain that one is immune from its influence, isn't it obvious one must undertake some form of blind test?

 

[Phelonious Ponk]

WRT expectation bias, it certainly can be the case that one can be immune from its effect in a comparison undertaking. OTOH, if one wants to be certain that one is immune from its influence, isn't it obvious one must undertake some form of blind test?

Yes, in that world where we rule out nothing and assume that anything is possible, someone can be immune to expectation bias. We probably have better odds of being shot out of a cannon and landing on the moon than we have of meeting this wondrous person, but it's possible. And until we were shot out of cannons at the moon, it was still possible that the world was flat. Of course that could all be a hoax. The world could be flat. And expectation bias could have a vaccine made of sophisticated taste and superlative equipment. But I'm not betting my retirement funds on it.

Tim

 

[Gregadd]

The lab allows us to control variables. Unfortunately the end product requires that we bring all those variables back for the real experience.

So my expectation bias was jacked through the roof until reality came crashing down when the needle hit the groove and I thought I was standing at the ocean during high tide. My expectation bias took an ass-whooping last night.

I guess my point is that expectation bias can’t fool me into thinking something that is clearly not true

I think I have been reminded repeatedly that blind testing is only required for small differences or differences that seem to deny known science. Moreover blind testing would only apply to external stimuli.

It would appear that what Frank heard was neither subtle nor defying science. I have also said that we should not "generalize to the specific." Therefore one anecdote would not constitute scientific proof.