The fallacy of Expectation Bias

I think some will use anything to support their argument. That includes accusing others of bias.

A few articles on the theory of expectation bias...

http://scholar.google.com/scholar?q=biased+expectation+theory&hl=en&as_sdt=0&as_vis=1&oi=scholart

A few articles on the psychology of expectation bias...

http://scholar.google.com/scholar?q=expectation+bias+psychology&hl=en&as_sdt=0&as_vis=1&oi=scholart

A few articles about clinical trials related to expectation bias...

http://scholar.google.com/scholar?q=expectation+bias+clinical+trials&hl=en&as_sdt=0&as_vis=1&oi=scholart

I know, of course, you believe that audio is special and because I haven't pointed you directly to any studies involving expectation bias and audiophile equipment, you'll discount everything here. We'll just have to live with that, as how audiophile hobbyists justify the inaudible differences between the things they pay good money for is not really on anyone's research radar. Regardless, the fact that you heard the noise on a really crappy pressing, in spite of your hopes doesn't delegitimize expectation bias.

Tim

mep said:

Tim-I don’t doubt that expectation bias is a real phenomenon in lots of areas of our lives. I just don’t think you can fool yourself with regards to audio gear over the long haul unless you are in denial, deaf, or you are just basically a dullard.

Click to expand...

Except for digital lovers as HP once said...."stare into my laser, I am perfect."

mep said:

Some people on this forum like to talk about the effect of expectation bias and the result it has on your perception of sound. The common *wisdom* is that if you think a new purchase is going to sound good (based on looks, cost, perceptions, reviews, etc.), it will sound good and meet your expectation bias.

Click to expand...

That is not what it says Mark. It says that it *can* make you think something is better than it is. It does not assure that is the fact as you found out. Indeed, if I play 64k MP3 against the CD and lie to you that the former is a high-resolution audio format better than the CD, you will never be fooled. Gross differences will overcome expectation bias.

It all gets tricky as differences become smaller. To wit, we actually don't know that you were right to declare the new LP inferior to what it was supposed to. You have shared a subjective experience which we appreciate but unfortunately we don't know if it really equals the truth.

The only way to rule out expectation bias is to sit you in a chair, blindfold and gag you, and then switch your components back and forth and see if you can pick out your new component that stole your expectation bias heart.

Click to expand...

That is the only tool we have. It is not perfect. And it may wipe out differences that might be there but we know that it removes bias. Since we rather be right conservatively than wrong catastrophically, such methods are used for audio evaluations .

I think if you’re the least bit intellectually honest with yourself and you aren’t deaf as a post, expectation bias doesn’t have an effect on the final outcome of how you perceive something to sound.

Click to expand...

Then I must be deaf as a post or dumb . When we acquired Pacific Microsonics, the makers of HDCD, one of their engineers gave me a quick test of HDCD vs CD version of the same. I swore to him that the HDCD version sounded so much more open, had more detail, etc only to have him tell me that he was playing the CD version both times! After he told me this, I took the test again and this time they both sounded the same. You can say that expectation bias of them being the same wiped out the imagined differences .

I think expectation bias can certainly lead you to believe that a new gadget is going to sound great, but in the end it can’t fool you into thinking that something that is inferior is superior.

Click to expand...

If the differences are small, it can certainly do that. Placebo effect > small differences in audio.

BTW, the opposite of your experience can also be true. You can be expecting something to sound worse, only to think it sounds better where in reality it was the same or worse.

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

Click to expand...

The only way to get rid of this belief is to be shown to be an audio fool in many situations and only then, you will believe that your ears+brain can be fooled unfortunately. There is no getting around this. I wish this was not the case. But countless personal and professional experience tells me it is.

I once participated on a blind test on another forum. I was part of the minority voting one way. The person conducting the test said I was wrong. I showed him he was because he had made a mistake and two of the files were identical yet most people had voted them different! I was him my home turf evaluating these compressed files. So I had an advantage. Regardless, all those people saying those two files sounded different even though they were bit for bit the same, were subject of some sort of expectation bias, that there "must be a difference." So they heard one.

The moral to the story is never listen to your expectation bias because the little bastard lies all the time.

Click to expand...

Unfortunately we don't know when it does, and when it doesn't.

mep said:

Tim-I don’t doubt that expectation bias is a real phenomenon in lots of areas of our lives. I just don’t think you can fool yourself with regards to audio gear over the long haul unless you are in denial, deaf, or you are just basically a dullard.

Click to expand...

Over the long haul? I haven't seen any studies of the endurance of expectation bias, but I'd expect you're right, regarding audio and just about anything else. Someone recommends a wine. It has a 5-star tag at the wine shop. The shop keeper tells you it's one of their finest Pinots. You pay $75 for the bottle. You take it home and convince yourself its really, really good. Happens all the time. Three bottles and six months later will you have noticed that, after all, you really like that $15 Pinot you drink every week better? I expect (my bias) that happens. But what happened to you this week is you ordered the fine wine (or vinyl) that you expected to be really good and it tasted like vinegar. Your expectations didn't have a chance of overcoming the quality issues.

But it hasn't a thing to do with the "fallacy of expectation bias," or with audio being in any way immune. It has to do with the extremity of the example.

Tim

mep said:

So would that be reverse expectation bias? I thought expectation bias was predicated on results meeting expectations, but maybe that was my expectation bias of the meaning of expectation bias. I’m trying to come to grips with the concept of expecting something to sound worse, yet you think it sounds better, but yet it really could be worse and your initial expectation bias was correct but you weren’t smart enough to recognize how smart your initial prediction was.

Click to expand...

My expectation bias was that I would completely 100% understand this entire paragraph.

The question of course is this really "expectation bias. That is to say could one point to reasons why a well recorded 45 rpm 200g carefully produced and stamped record should sound better than an inferior reproduction. Expecting a Ferrari to perform better than its inferior competition is not bias. It's a valid decision. If the Ferrari does not live up to it's reputation than something must have happened.
Was your expectation invalid?
Was the Ferrari malfunctioning?

I do not recall the mention of any A/B test. Of course the argument usually is you did not hear a difference because there is none or that you heard it because you wanted to.

I suggest either Frank had a maifunctioning set of of records or he wanted a sound that a 45 rpm 200gram record could not deliver. My money is on a malfunction.

Hi

On one side is a system with your favorite system and cable and playing your favorite music .. System cost say 125,000 ( I like that number ) On the other side a system from component you don't know well but consider nonetheless middle of the road audiophiles, say NAD amps, PSB speakers but the price is 25,000.. Rig is RTR in both cases .. Very honestly which one do most people , audiophiles or otherwise expect to sound better? Which they're likely to prefer?

FrantzM said:

Hi

On one side is a system with your favorite system and cable and playing your favorite music .. System cost say 125,000 ( I like that number ) On the other side a system from component you don't know well but consider nonetheless middle of the road audiophiles, say NAD amps, PSB speakers but the price is 25,000.. Rig is RTR in both cases .. Very honestly which one do most people , audiophiles or otherwise expect to sound better? Which they're likely to prefer?

Click to expand...

Bias suggests that a belief in a certain outcome that is based on a prejudice or on factors that do not contribute to the sound.

If the increase in price etc. is invested in improvements that result in better sound. Most equipment has compromises that bring them to a certain price point

I’m not sure I understand your question. Do you question why a record recorded at 45 RPM over four sides should sound better than the same album recorded at 33 RPM over two sides?

Click to expand...

What I am saying that if executed properly the former should sound better(especially the inner groves). But that belief is not the result of bias. It is a perfectly valid conclusion The question is whether the changes were executed properly. Furthermore were the changes you expected a necessary result of those changes.

Greg

Your reply does indeed reveal a bias ...

FrantzM said:

Greg

Your reply does indeed reveal a bias ...

Click to expand...

Perhaps I did not state it clearly enouh for you. Would you care to elaborate?


Wiki
A cognitive bias is the human tendency to make systematic decisions in certain circumstances based on cognitive factors rather than evidence. Bias arises from various processes that are sometimes difficult to distinguish. These processes include information-processing shortcuts, motivational factors, and social influence (Wilcox, 2011). Such biases can result from information-processing shortcuts called heuristics. They include errors in judgment, social attribution, and memory. Cognitive biases are a common outcome of human thought, and often drastically skew the reliability of anecdotal and legal evidence. It is a phenomenon studied in cognitive science and social psychology.
[edit] In the media