How do we end up knowing things that just aren't so?
The brain is hard-wired to detect order in the nature of things. We can learn from experience by accumulated observations and this has obvious survival advantages in evolutionary terms.
But where do things start to go wrong? First of all, we see ordered patterns of outcomes that are in fact the blind product of chance. Chance produces less alternation than our intuition leads us to expect. If we toss a coin 20 times, we're unlikely to see 10 heads and 10 tails. A series of 20 tosses has a 50-50 chance of producing 4 heads in row. When we see patterns such as the lucky streak in baseball we think we are spotting an order that isn't in fact there.
The regression effect also fools us into misattributing a cause to an effect. You perform exceptional badly or excel when taking an exam, much worse or better than your average. Your next result is likely to be better or worse as you move back to your average. That's the regression effect. But we make assumptions that the exceptional and atypical is representative when the regression effect would tell us otherwise: investors may assume that a company's bumper profits in one year will predict will be repeated in future years when in all likelihood they will actually fall.
We underdetermine beliefs with insufficient data, treating weakly tested hypotheses as facts. We look for confirmatory examples while overlooking or discounting facts that contradict a belief. We fail in other words to understand the distinction between necessary and sufficient evidence. We seize on isolated, salient examples of pieces of data that prematurely confirm a hypothesis. Take the homoeopathist's claims that a cancer patient was miraculously cured after taking an alternative remedy. The recovery is treated as conclusive evidence of the remedy's efficacy. But such evidence is in itself insufficient to proof anything - isolated facts do not in themselves provide sufficient confirmation. They are too vulnerable to the discovery of counter-examples that contradict the hypothesis.
We leap to such conclusions because when we test for a hypothesis, we fail to define what success or failure is. Too often beliefs are formed with vague definitions of what counts as a successful confirmation. Studies of identical twins separated at birth may well track an identity of life outcomes that point strongly to genetic influences. But there are many outcomes or results in any given life. Some of these may overlap and give the impression of congruence. So the twins may both choose the same occupation and this is indeed a striking identity of outcome but this is only one such outcome, and others may vary. The danger once again is taking an overlap of outcomes from two sets of data similar while overlooking variances. Likewise many predictions are couched so vaguely to guarantee against disconfirmation, akin to Woody Allen's spoof Nostradamus character who portentously avers that `two nations will go to war but only one will win'.
Does our social nature compensate for this? Not necessarily. We tend to associate with like-minded people and to fight shy of conflict and controversy. Therefore members of presidential advisory groups hold their own counsel. We keep our mouths shut during a meeting at work. We do not want to be seen to rock the boat. The result is that others believe that their beliefs are more broadly shared than they actually are (one reason why the bore and the name dropper carries on with a self-defeating strategy is precisely the reluctance of others to point it out).
Good heavens, having said all this how on earth can we tell if our beliefs our well founded? There is no easy way out of these cognitive illusions. It's not all bad. We do have good reasons for example to accept the theory of gravity, which has weight (so to speak) and well attested by centuries of sense and statistical data. So we can rightly disregard claims of levitation on this basis.
We can also tighten up our definitions of what counts as confirmation, as we noted earlier. If we were testing whether a training course that claims it can raise sales staff performance really works, then we would define successful confirmation as increased sales figures. The scientific process of peer review also helps: we can make sure that a researcher does not which members of the trial group are receiving the new drug being tested, so preconceptions of success or failure do not contaminate the researcher's observations. We can test if a claim for an extraordinary effect like Extra Sensory Perception can be replicated (it can't).
These are palliatives however. We can only strive imperfectly to try and recognise when our reasoning faculties are leading us up blind-allies. This book will help you at least be a little more vigilant when it comes to forming conclusions about why you think you are right to believe the way you do
Download the free Kindle app and start reading Kindle books instantly on your smartphone, tablet or computer - no Kindle device required. Learn more.
Read instantly on your browser with Kindle Cloud Reader.
Using your mobile phone camera - scan the code below and download the Kindle app.
Flip to back Flip to front
Follow the author
Something went wrong. Please try your request again later.
| Amazon Price | New from | Used from |
|
Kindle Edition
"Please retry" | — | — |
|
Audible Audiobooks, Unabridged
"Please retry" |
£0.00
| Free with your Audible trial | |
Black History Month 2022 - Editors' Picks
Frequently bought together
- +
- +
Total price:
To see our price, add these items to your basket.
Choose items to buy together.
Customers who viewed this item also viewed
Page 1 of 1 Start overPage 1 of 1
About the author
Follow authors to get new release updates, plus improved recommendations.
Customer reviews
Top reviews from United Kingdom
There was a problem filtering reviews right now. Please try again later.
Reviewed in the United Kingdom 🇬🇧 on 20 November 2011
Reviewed in the United Kingdom 🇬🇧 on 16 October 2008
Reviewed in the United Kingdom 🇬🇧 on 11 January 2021
Reviewed in the United Kingdom 🇬🇧 on 30 November 2010
