Monday, February 26, 2007

The Power of Critical Thinking: Bias and Self Deception

Bias:

Some time ago, during one of those post-conference informal discussions, I heard a fellow give the best description of bias that I've ever heard. I will attempt to paraphrase:

Bias is the natural result of every person's self-centrism. We all - except for the few who are incapacitated by self-doubt - are utterly convinced that our beliefs, actions and opinions are absolutely correct at the time.

The caveat "at the time" is an essential part, because all of us (except for those who are pathologically incapable of acknowledging error) have had occasion to realize that we have made a mistake, that our beliefs, actions and/or opinions were wrong.

However, at the moment we thought that, we were convinced we were correct - correct about us having previously been in error.

You can see how this could rapidly get out of hand.

But it holds together very well as a general statement of how people view their own thinking process. We (with very few exceptions) are sure that we are doing or thinking the "right" thing at the time that we are doing it or thinking it.

So, what has this got to do with bias? Plenty.

Bias is a sort of "blind spot" in a person's thinking - a place where their assurance of being right makes them vulnerable to imagining the world to be different from how it truly is. It is, in short, a minor delusional state.

As Mark Twain is reported to have said: "It ain't what you don't know that gets you into trouble. It's what you know for sure that just ain't so." So it is that the blind spot of bias gets people into trouble. Ignorance is merely a need for information or education; error - especially when it is believed wholeheartedly - is a calamity waiting to happen.

A good analogy would be a map. If your map of reality has blank spots, you are likely to be more cautious in those areas. You'll ask questions, listen to what people who have been there have to say and keep you eyes open. On the other hand, if your map has roads where there are none and smooth plains where there are cliffs and pits, then you are standing into trouble.

Bias can lead to self-deception (more about that later) when it convinces a person to ignore good information because it conflicts with their pre-formed (and self-corroborated) view of reality. Some spectacular examples of this have occured in the media in recent years, when otherwise sage and seasoned journalists were completely taken in by fakes and phonies. That they often did so in the face of evidence that they were being hoodwinked only serves to underscore the pernicious nature of bias.

As an aside, it is laughable in the extreme to see people state that they are "not biased". Ridiculous! Everybody is biased in some way - everybody has an opinion on everything, even if that opinion is "That's too unimportant to think about." As a result, everybody is prone to - biased toward - a particular view of reality.

To reiterate: no matter what you think, you think that's the right thing - at that time. If it were a conscious decision, it wouldn't be bias. Bias influences, shades and slants conscious decisions, in ways that we are not aware of at the time. Looking back, we may be aware of how our assurance of correctness led us to disaster, but we don't - we can't - see it at the time.

To be sure, people are often aware that they are doing or thinking things that "aren't right", but that is the imposition of an external measure of correctness. It has nothing to do with whether the person feels that their thoughts or actions are right for them, in their situation, at that time. The person who is shoplifting a loaf of bread knows that it is a crime (possibly even a sin), but is assured that they are doing the right thing for their own reasons.


Self Deception:

Bias is usually a necessary precondition for self-deception. It would take an incredible strength of will to truly believe something you know to be false. I'm not sure that it can be done this side of insanity.

However, humans have shown themselves to be masters of convincing themselves that the real is false and the false is real. And the first step in this process is to be assured of the correctness of one's own thinking - our old friend bias!

In some ways, it's rather amazing that people can ever avoid self-deception. What keeps most people sufficiently grounded in reality to allow them to carry out the day-to-day activities of living is a regular contact with aspects of reality that cannot be easily ignored.

Someone who has convinced themselves of their ability to fly (sans aeroplane) will rapidly (at 9.8 m/sec/sec) be disabused of this notion. They may even survive to "internalize" the lesson. Likewise, reality intrudes its unwanted self into most self-deceptions, sometimes, sooner, sometimes later; sometimes subtly, sometimes dramatically.

But there are some self-deceptions that are resistant to reality therapy. Some of them lack a sufficient grounding in reality to ever run across a contradiction. Most religious thought is of this nature. There is simply not enough contact between religion and reality - with some notable exceptions - to provide a convincing "whack" to the deceived... er, the devout.

Other times, the collision with reality is a long time coming, allowing people a "grace period" to believe (and perhaps entrench) their self-deception. For example, people who try quack remedies for real illnesses often have a period of time before the reality of their unchecked disease breaks through their belief. Even worse off are the people who try quack remedies for imaginary illnesses - they will now have two mutually-reinforced self-deceptions!

Additionally, there are some counter-reality thoughts that can diminish the impact of reality. Some of these are (not an exhaustive list by any means):

- It would be so embarrassing if I were wrong!
- I've invested too much time/money/effort/reputation on this to admit that I was wrong!
- People would be so mad at me if I were wrong!
- I really trust the person who told me this...they can't be wrong!
- I really need for this to be right!


You'll notice that I haven't included anything like "It's all a conspiracy!" or "They're lying to me!" That's because these sort of thoughts are the result of self-deception (and might even be considered diagnostic of it) rather than contributing factors. Once a person has convinced themselves that the cost of being wrong is (for them, at this time) greater than the cost of persisting in error, then the mind will generate a suitable set of excuses (sometimes called "rationalizations" or even "delusions") in order to maintain their denial of reality.

Rationalizations are a way of covering up the gap between reality and a person's conception of reality. The more their mind-generated world-view comes in conflict with reality, the more rationalizations are needed. At some point, they may even cross the imaginary line between denial and delusion.

People who are in denial are often very hostile toward people who try to bring them back to reality. After all, it's hard work to maintain all those rationalizations and they do not appreciate visitors who want to track reality all over their snug, safe sanctuary. They will often lash out at people who have the temerity to disagree with them.

As a result, people in denial often seek out others with the same - or similar - world-view. The Internet has aided this process immensely, providing innumerable places for people to gather to share, refine and reinforce their denial of reality.


Avoiding Self Deception:

While it would be nice if we could all look reality in the face at all times, this simply is not the human condition. Failing a complete refit of the human psyche, what can we do to avoid self-deception?

[1] Check your ideas with someone else.

This is the basic concept behind "peer review" in scientific journals. Your ideas, methods, data and conclusions are put before a number of independent (they aren't all sitting in one room, influencing each other) reviewers who read it and give their critique (and usually criticisms). This provides a number of minds that - in all probability - do not have the same biases (the same blind spots) as the author(s).

For the average citizen, "peer review" can be a bit harder to find. A common mistake is to ask someone who already thinks as you do to critique your idea. Thus we have UFO conspiracists asking other UFO conspiracists if the small sharp thing in their bum is a splinter or an alien mind-control device, with the predictable bizzare answer.

This is also where a lot of the news media have fallen foul of self-deception. In a group with the same political and social mindset, it may be hard to find someone to say, "Gee, Dan, that memo doesn't sound very believable to me. Maybe you ought to check it out better." It's also unreasonable to expect people to tell the boss that they think he's gone crackers. Better to find an independent appraisal.

[2] Be skeptical of everything you hear, especially if you agree with it.

Again, the news media would have saved themselves a number of black eyes and bruised egos if they had followed this simple rule. For that matter, a number of people in science would have been better off if they, too, had heeded this advice.

You are very much more likely to be taken in by a falsehood that conforms to your world-view (your biases) than to believe a truth that doesn't. Keep that always in your mind. Question the basis for any statement that claims to be "fact"; question it twice if you find yourself wanting to believe it.

[3] Occasionally step back and ask yourself, "What would it take to make me believe/not believe this?"

When we are self-deceived, the answer to this simple question - if we are honest with ourselves - is most often "There is nothing that would change my mind." This is a very bad answer because it means that your belief (or non-belief) in something is a matter of religious devotion, not reason. If that's OK for you, so be it. But it might just prevent a nasty misstep if you can recognize that you are not seeing the world as it really is.


This has just scratched the surface of a topic that could fill volumes (and has!). Perhaps I'll add another volume to the world's collection someday. It's on my list (right after cleaning out the attic).

Until next time!


Prometheus

      17 Comments:

      Blogger mcewen said...

      '[2] Be skeptical of everything you hear, especially if you agree with it.' - I like this one best. I thought it didn't apply to me because I'm a cynic and a pessimist, but of course that means that I've already taken a position!
      Now I'll just take the plank out of my own eye......
      Cheers

      26 February, 2007 17:03  
      Blogger Do'C said...

      Excellent article Prometheus. I love that Mark Twain quote.

      26 February, 2007 18:39  
      Blogger Camille said...

      When I was a research assistant for a social psych professor at UCD I was helping do research in bias.

      It's possible to be made aware of a bias but not be able to rid oneself of it's unconscious effects, thus one can tell oneself that there's no reason to fear persons of a certain race or gender or age, or to think they are weird, or whatever one's bias might lead one to think... but the bias does not go away. It will still influence the actions and words of the biased person.

      The sad thing is that some of the biases come from the media, where one can be constantly "trained" to believe that a stereotype is TRUE. Constant repetition of stereotypes wires the brain up in a way that links, say, "old people" with "bad driver" or "African people" with "live in a mud hut" or whatever.

      Once the biases are engrained it takes lots of repetitions of things like pictures of Africans living in cities, using computer, driving cars... to overcome the old bias. Some biases never go away.

      Interestingly, people with poor executive function (like me) are more likely to be aware of a bad bias but to stumble all over themselves when the bias is put to the test, like I might say, "Oh, you're A JAP-AN-NEEZE. I LOOVE YOUR PEOPLE'S SOOOSHI." Well, I wouldn't say that one, but some of that kind of stupid stuff is more likely to come out of the mouth of a person who is aware of their bias and trying to overcome it but isn't good enough at controlling their thoughts in real time due to executive function problems.

      The bias against handicapped people is amazing. No wonder no one wants to be related to a handicapped person, much less to be responsible for creating one, at least in the majority US culture now.

      26 February, 2007 19:59  
      Blogger Joseph said...

      - It would be so embarrassing if I were wrong!
      - I've invested too much time/money/effort/reputation on this to admit that I was wrong!
      - People would be so mad at me if I were wrong!
      - I really trust the person who told me this...they can't be wrong!
      - I really need for this to be right!


      I think there's someone (*cough* Blaxill *cough*) who is going through this, and prefers to keep quiet about it.

      27 February, 2007 05:12  
      Blogger Camille said...

      "Interestingly, people with poor executive function (like me) are more likely to be aware of a bad bias but to stumble all over themselves when the bias is put to the test, like I might say,"

      I wrote that wrong. I mean that people with poor executive function IF they become aware that they have a bias have a hard time dealing with consciously overriding the bias.

      We can consciously override a bias. As you pointed out in your blog. Say if our bias is that all people of a certain ethnicity would make bad employees, if we are hiring people, we can tell ourselves that our bias is misleading (or that it's against the law to act on it) and go ahead and hire a person of that ethnicity. It doesn't mean that the bias is gone, it's still there.

      27 February, 2007 12:06  
      Blogger Prometheus said...

      Camille,

      Bias is a hard thing for anyone to detect, since we all think that it is correct. Often, when it clashes with other aspects of our world - such as when a colleague brings it to our attention - we can become aware of it. What people do with that information is an interesting insight into their personality.

      Joseph,

      I haven't heard much from Mr. Blaxill since his appearance as a "sandwich author" in the paper by Herbert et al in Neurotoxicology (Sept 2006). Has he made any statements since then?

      I expect to see some of the more prominent figures in the autism-mercury world to start their move away from thimerosal and toward some other "toxin" (perhaps mercury in the environment). It may be a conscious move on the part of some, but most will simply be moving away from the increasing cognitive dissonance of having their beliefs grind against reality.


      Prometheus

      27 February, 2007 17:19  
      Blogger Fore Sam said...

      Prometheus, You have a good point here. I always think I'm right when I bet on a horse but more than half the time, I'm proven to be wrong. It's easy to admit this because some of my money is gone. The rest of the time, I am right and I have more money. I proved i was right about chelation when my son began to improve. If you could watch your kid improve by trying chelation, you would prove your bias to yourself. You can't win if you never make the bet.

      27 February, 2007 18:14  
      Blogger Prometheus said...

      OK, does anyone else see the flaw in Fore Sam's thinking? Yes, you in the back?

      Absolutely right, a person who has invested time, money, and reputation into the outcome of an experiment is often biased in their expectations and will interpret any data according to that bias.

      Fore Sam has invested time, money, effort and credibility into proving that chelation can "cure" autism and ADD/ADHD. It would be terribly difficult for him to admit that the changes that he's seeing are not necessarily due to the effect of chelation.

      On the other hand, I have always maintained that I am open to the data on chelation and autism. I'm still open. I just haven't received any data.

      Kids with autism, like kids with most developmental disabilities, continue to progress and improve. As I have said on innumerable occasions, autism is a syndrome of developmental delay, not developmental stasis. Seeing progression or improvement in a single child - or even a small group of children - is not proof of chelation's effect.

      Once again, here is how you should study chelation. And once again, it is not my responsibility to do the studies or gather the data. You're trying to convince me, not the other way around. I have absolutely no interest in trying to change Fore Sam's mind about chelation - what he does with or to his children is between him and the local child protection services.

      To adequately study chelation, you need to make sure that the effects you see are due to the chelation, not the natural development of the child. Since autism is a poorly defined disorder, the rate of development ranges widely, so a large number of children will need to be studied - a couple hundred would do for an intitial study, a couple dozen for a "pilot study".

      Randomly assign the children (by lottery) to either the chelation group or the non-chelation group. You don't necessarily need to use a placebo (since it might be hard to find a convincing placebo), but the people evaluating the children's development must not know which group the child is in. That means that parental reports, checklists and surveys cannot be used.

      The only way that parental reports could be used is if a convincing placebo could be developed. And you would have to test that by seeing if the parents could tell which group their child was in.

      This is all pretty much standard scientific method for clinical trials. Which is why I am so puzzled that nobody has done it yet. Or, maybe they have and decided they didn't like the results.

      After all, the folks providing chelation have all the business they can handle, so why should they risk the goose that lays the golden eggs by studying it too closely. As long as parents flock to chelation without asking for data (not testimonials) about how well it works, the chelationistas have absolutely no reason to collect any data - it can't help them and could only hurt them.

      So, as I've said before (and will doubtless say agin), I am willing, ready and able to be convinced. The only way to show that I'm biased is to provide me with the data (as I've outlined above) and then see what I do.

      All I'm asking for is a little data. How hard can it be to put together a little pilot study of a couple dozen kids? But nobody does - which makes me suspicious.

      Only a fool bets when they don't know what the adds are.

      Prometheus

      28 February, 2007 08:46  
      Blogger Joseph said...

      Jim Adams was supposed to report on preliminary results of a double-blind study of chelation by the end of 2006. I'm not sure why he hasn't. The study turned out to be very small in the end as I recall, with only 12 participants left or something. Maybe it fell apart competely. Maybe he doesn't want to say anything about the results just yet. He had promised he would announce results regardless of what they were, as he should.

      28 February, 2007 09:08  
      Blogger Fore Sam said...

      The odds are, sticking the kid in an institution if he isn't cured or watching him enjoy his life if he is. If it doesn't cure him, the institution will be waiting. How can you lose?

      28 February, 2007 15:26  
      Anonymous anonimouse said...

      Absolutely right, a person who has invested time, money, and reputation into the outcome of an experiment is often biased in their expectations and will interpret any data according to that bias.

      It's sort of like the Concorde effect - I've seen it all the time in business. Once you've sunk a lot of money and time and effort into something, you are more inclined to continue to spend money to try to "salvage" any positive results, no matter how one-sided the cost-benefit ration may be or how minimal those results really are.

      So it doesn't surprise me that John continues to pursue chelation therapy. Someone with a gambler's personality like Best fits that bill to a T.

      He had promised he would announce results regardless of what they were, as he should.

      I'm sorry, that made me chuckle.

      01 March, 2007 05:48  
      Blogger Prometheus said...

      Fore Sam,

      As I've said many times before, don't ever change. I couldn't pay someone to feed me the lines the way you do!

      What Fore Sam has done is set up a false dichotomy, an "either-or" choice that doesn't really exist.

      To begin with, there are a lot of other potential outcomes to autism besides the two he has tried to restrict us to (institution or cure). As they have for the decades that they've been studied, autistic kids grow up to be adults with a wide variety of abilities, even without chelation. Some will be able to live on their own, others will need constant care. What is for certain is that not all of them will need to be put into an "institution" - even if they aren't chelated.

      Fore Sam would like us to believe that the inevitable outcome for an untreated autistic child is to be placed in an institution. This, fortunately, is not the case - and never was. Temple Grandin is not in an institution, nor are a large number of autistic adults. Even if Fore Sam tries to pretend they don't exist.

      Now, let's examine the real choices. Given that the range of outcomes without treatment range from successful, independent adult (for instance, one with a PhD and worldwide speaking engagements) to institutionalized adult, what sort of data can Fore Sam provide that shows chelation can increase the chance that an autistic child will grow up to be an independent adult?

      None.

      That's right, he (and the rest of the chelationistas) have not presented any data to support their claim that chelation improves the outcome for autistic children; short-term, long-term, any-term.

      Now, Fore Sam will doubtless point to the numerous testimonials (including his own) about how chelation has "cured" children. However, I can counter with testimonials about children who "recovered" without any chelation. But we don't need testimonials in this situation - what we need is data.

      So, can Fore Sam provide any data that shows children who receive chelation either show more improvement or improve faster than children who don't get chelation?

      No.

      Or, at least, he hasn't chosen to. Nor has anyone else.

      I'm still waiting for the data on chelation and autism that was promised back in 2004. I suspect that I'll be waiting for a long, long time.

      And the longer I wait, the more I begin to suspect that there isn't any data. I'm trying to not be cynical, but it's hard.


      Prometheus

      01 March, 2007 20:40  
      Blogger Joseph said...

      Very few classic autistics end up in actual institutions or mental hospitals. It's something like 5%. Most seem to end up living in community care or with parents. The number of adult classic autistics who live independently is small, but it seems to be a bit higher than that of those who live in institutions.

      02 March, 2007 07:45  
      Blogger David said...

      "Or, at least, he hasn't chosen to."

      Indeed.

      He was offered a chance to do a case study... never conclusive, but nonetheless useful, on his son... He practically told me to fuck off.

      02 March, 2007 13:20  
      Blogger Prometheus said...

      David,

      I'm pretty firm in my belief that advocating a point of view does not require that one expose themselves or their families to intrusion or abuse. I respect Fore Sam's right to keep his son's medical records and other personal data private. That said, anything he cares to put forward as data, evidence or proof is fair game for discussion.

      I would like to reiterate that simply by choosing to participate in a public discussion of a controversial topic should not be construed as a license for abuse and harrassment. I'm speaking specifically about the abuse and harrassment that has been directed toward Amanda, of the Ballastexistenz blog.

      If there are still adults in the world who think that it's appropriate to harrass people who disagree with you or who - God Forbid! - want to tell their side of the story, I have a message for you:

      It's not appropriate!

      For people who need their clues a bit more direct, the translation would be:

      Knock it off! Grow up! Show a little common decency for a change!

      That is all.


      Prometheus

      02 March, 2007 14:12  
      Anonymous anonimouse said...

      If Fore Scam wants to trot his son as an example of a child "recovering" through use of chelation therapy, he should put his gambling money where his mouth is. Otherwise, it's just the bluster of a paranoid black-helicopter watcher who vacillates between hating the government and wanting them to give him money for his defective kid.

      You don't get to have it both ways. If you want to make the claim that chelation is curing your kid (an extraordinary claim) you'd better be prepared to prove it.

      05 March, 2007 15:28  
      Blogger Prometheus said...

      Anonimouse,

      As I said, if someone (e.g. Fore Sam) brings in personal or family information (e.g. the "cure" of a child), then that information is fair game for discussion. However, having brought that information to the discussion does not obligate them to provide any and all information that the group needs/wants to assess the claim.

      Mind you, if the claimant fails to provide the data necessary to support their claim, they have only themselves to blame if we are not convinced. And they can't use the "it's private information" gambit to assert their point - if they can't or won't provide the data, their claim fails.

      We can't be expected to simply take someone's word that they have the data; we've been - I've been - left waiting years (and still waiting) for data that's "just around the corner" or "will be published soon". Either you put up or - in an ideal world - you shut up.

      If the data is too private to share, then you'll just have to be satisfied with having at least convinced yourself.

      In the end, I won't expect anybody to share personal or private information in any discussion. If they choose to share such information, they are free to decide when they've shared enough. And we, in turn, are free to decide whether or not the data we've been given is sufficient to establish their claims.

      It's as simple as that.

      A word of caution to those who feel the need to share personal or family stories: your story, no matter how moving or heart-rending it may be, is just one story - no more, no less. It is an anecdote, not data.

      A very wise biologist once told me that, in medicine and biology, einmal ist keinmal (once is never). By that, he meant that anything can be seen once - it's the ability to repeat the observations that makes them real and not artifact.


      Prometheus

      06 March, 2007 08:48  

      << Home