Ben Goldacre, The Guardian, Saturday 4 December 2010
Why do clever people believe stupid things? It’s difficult to make sense of the world from the small atoms of experience that we each gather as we wander around it, and a new paper in the British Journal of Psychology this month shows how we can create illusions of causality, much like visual illusions, if we manipulate the cues and clues we present.
They took 108 students and split them into two groups. Both groups were told about a fictional disease called “Lindsay Syndrome”, that could potentially be treated with something called “Batarim”. Then they were told about 100 patients, slowly, one by one, each time hearing whether the patient got Batarim or not, and each time hearing whether they got better.
Now, when you’re hearing about patients one at a time, in a dreary monotone, it’s hard to piece together an overall picture of whether a treatment works (this is one reason why, in evidence based medicine, “expert opinion” is ranked as the least helpful form of information). So while I can tell you that overall, in these results, 80% of the patients got better, regardless of whether they got Batarim or not – the drug didn’t work – this isn’t how it appeared to the participants. They overestimated its benefits, as you might expect, but the extent to which they overestimated its effectiveness depended on how the information was presented.
The first group were told about 80 patients who got the drug, and 20 patients who didn’t. The second group were told about 20 patients who got the drug, and 80 patients who didn’t. That was the only difference between the two groups, but the students in the first group estimated the drug as more effective, while the students who were told about only 20 patients receiving it were closer to the truth.
Why is this? One possibility is that the students in the second group saw more patients getting better without the treatment, as so got a better intuitive feel for the natural history of the condition, while the people who were told about 80 patients getting Batarim were barraged with data about people who took the drug and got better.
This is just the latest in a whole raft of research showing how we can be manipulated into believing that we have control over chance outcomes, simply by presenting information differently, or giving cues which imply that skill had a role to play. One series of studies has even shown that if you manipulate someone to make them feel powerful (by remembering a situation in which they were powerful, for example) then they imagine themselves to have even greater control over outcomes that are still purely determined by chance, which perhaps goes some way to explaining the hubris of the great and the good.
We know about optical illusions, and we’re familiar with the ways that our eyes can be misled. It would be nice if we could also be wary of cognitive illusions that affect our reasoning apparatus, but more than that, like the “close door” buttons in a lift – which, it turns out, are often connected to nothing at all – these illusions are modern curios.
QuietKnoll said,
December 3, 2010 at 10:41 pm
I suspect this might explain why many “successful” business people believe that their hard work – rather than luck in gambling with the vagaries of market systems – is the cause of their success. Sadly this comforting self delusion is then forced onto public services as an evidenced based approach to policy making instead of being seen as faith based mumbo jumbo.
ken said,
December 4, 2010 at 5:43 am
Even worse are doctors who say “I had a patient with similar symptoms the other day and I prescribed..” I wonder how many doctors look at Cochrane?
QuitetKnoll – more (apparently successful) business people believe that it is their genius rather than hard work. The true ranking I believe is hard work/luck/genius.
Regression To Norm said,
December 4, 2010 at 7:15 am
I cannot agree with QuietKnoll’s assessment that the experts who look after the vast amounts of money invested in stocks and shares around the world merely rely on luck. They clearly have greater insight than he can possible understand. Obviously if it were just luck, and general trends, rather than skill and wisdom then every now and then everything would go horrible wrong!
Damien P said,
December 4, 2010 at 7:47 am
This is the same reason stock market dealers think they know what they’re doing when it fact they’re bascially guessing.
You get what’s called a Gaussian curve of performance. Some investors will do consistently better than average even if everyone is guessing. It’s like thousands of monkeys flipping coins; some of the monkeys will get 10 heads in a row. These are the monkeys that will be celebrated for their insight. And they themselves will believe that they have some special skills and talents that enabled them to get the 10 heads in a row.
Self delusion with shades of Dunning-Kruger too.
Damien P said,
December 4, 2010 at 7:53 am
Just realized QuietKnoll made the same point hours ago.
D’oh!
(Although to be fair I did get in a scientific sounding performance graph reference!)
fenderstrat said,
December 4, 2010 at 10:25 am
Isn’t this more an illusion of causality than an of control? I’m trying to gather a few good examples of illusions of causality. Thread in “Bad Science” on the boards if you have any ideas. Cheers.
Also – that lift example? True? Really? Link? Cheers again.
Moobs said,
December 4, 2010 at 11:35 am
If you want to see this in action (and can bear it) watch an episode of Deal or No Deal. A process which consists of non-random selection of boxes with unknown contents (thus producing near random output) is encrusted with with enough delusion of control and pseudo-spirituality that I wonder if we are seeing a religion in mid-formation. I saw it last night and a contestant was called on to “focus” when choosing (to what conceivable end?), then wandered around touching boxes mumbling “I believe in myself”. At the end she explained that her winnings had been “destined”.
Dean Morrison said,
December 4, 2010 at 11:46 am
The Wikipedia entry on the psychology of ‘Framing’ gives a similar example:
en.wikipedia.org/wiki/Framing_effect_(psychology)
iamjohn said,
December 4, 2010 at 1:49 pm
I now know much more about lifts than I ever thought I would – the danger of chasing links.
Gambling is another example of a cognitive illusion. I ‘accidentally’ spent about a month in Las Vegas a couple of years ago and was utterly enthralled by people’s gambling reasoning. One proud husband told me at length how expert his wife was in slot tournaments (where there is literally just one button to press repeatedly for 7 minutes), others about their systems, skills and of course ability to predict cards. his confidence in the illusion of control is confounded with the luck of a few and the very dubious memory of the rest.
bob_bura said,
December 4, 2010 at 6:14 pm
This study was also covered nicely by the BPS Research Digest last week:
bps-research-digest.blogspot.com/2010/11/can-psychology-help-combat.html
Dick Pountain said,
December 5, 2010 at 3:58 pm
Actually it’s my experience that a lot of very successful business people actually believe superstitiously in their own luck, just like gamblers. They only trot out the “hard-work” response in interviews because they think it’s more (neo-liberal) ideologically sound.
Dick Pountain said,
December 5, 2010 at 4:04 pm
Please ignore one of those “actuallys” – whichever you prefer.
tialaramex said,
December 5, 2010 at 4:38 pm
Yes, the modern world has plenty of buttons that “don’t do anything”. But they ARE connected, and the article linked makes that clear. The building’s engineers, as well as elevator technicians, and in rare emergency, firemen, can use this button. Ordinarily an unattended elevator’s doors must close automatically, or else the elevator quickly becomes useless, stranded with its doors open far from people who want to use it. Having set an optimal time for the doors to close based on sensors detecting people nearby, it doesn’t make sense to allow mere users to override it. But with the elevator locked to local control the situation changes – you can’t call the elevator from another floor in this mode anyway, so the “Close” button actually closes the doors.
At pedestrian crossings there’s a slightly different rationale. At some crossings all pedestrian phases are optional – if there are no pedestrians there’s no reason to stop the traffic. These of course have working buttons. When a pedestrian presses the button, the crossing enters “stop traffic” phase as soon as possible, then activates the crossing, then shuts it down and enters a “dormant” phase for a while (so that frequent pedestrian requests don’t shut the road altogether). But at other crossings the normal operation of the junction includes a phase when it’s safe to cross anyway. In this case the button isn’t needed, but it’s included because it’s reassuring to users who aren’t traffic engineers. Sometimes that “pedestrian safe” phase does not exist in a low-traffic mode used e.g. at night, in this case the button works when in that mode but not at peak times.
anon3455 said,
December 6, 2010 at 8:35 am
DP, I think the use the “hard work” response because they understand that represented that way there is no political pressure on changing the distribution of income. And of course its good for the ego. The ideology doesn’t factor in.
Hard work sounds positive, whereas gambling, as we all know, tends to cause all sorts of problems. So publically supporting the latter as a basis of economy is essentially telling people that the political left is right. Which is something that rich people think is bad, since it might raise their taxes somewhat – nevermind that they would still make obscene amounts of money that they pretty much have no use for.
BJ said,
December 6, 2010 at 1:26 pm
This seems to have moved to a lefty ‘why haven’t I got much money, whereas other, far less intelligent people have’ discussion.
Whilst it’s fair to say successful traders will have some luck in their careers, they also put in huge amounts of time & effort analysing the markets. Market’s didn’t move on George Soros’s opintions because he rolled dice.
youngfogey said,
December 6, 2010 at 3:01 pm
Thanks for this. I have always gained an immense sense of power seeing people who believe that pressing the lift button would make lift doors close quickly. Similarly, on trains, there are people who believe that the driver releases the door lock and if you lean out of the window he’ll see you. Not so. It’s automated. So too are seatbelt signs in planes. The pilot has more important things to do. Automatic doors! People believe that waving at the sensor makes them open and they have control, but it’s a fallacy. There’s nothing there. Red traffic lights have nothing to do with traffic flow but are controlled by a complex software system, all automated. Speed cameras do not stop working if you have a blue light on the front of your car. It’s all about people believing they have control. I know someone who believes that having a fast car means the police won’t be able to catch him. Not so. Average speed checks are everywhere. But once you know the innermost workings of the machine, it gives you a great sense of power.
QuietKnoll said,
December 6, 2010 at 3:10 pm
BJ: For me Soros is THE example that proves the rule (regarding Luck vs Hard work in financial dealing). He wrote that he looked for early changes in trends and then bet on the possibility of making a mint before the herd behaviour of the market woke up to the fact. He admits he got 9 out of 10 such predictions wrong and lost his bet. However he made so much from the 1 in 10 that worked out that this more than covered his losses.
Ken: In this regard Soros was a genius!
Sadly – especially for “Mom & Pop” investors who buy into bogus trading advise to help them be the next George Soros the automated trading programs that spot and then magnify trend changes are only adding to the Las Vegas casino atmosphere. Its the financial equivalent of homeopathy – genius diluted to the point of absurdity (i.e. luck!)
general_jumbo said,
December 7, 2010 at 3:08 pm
BJ: Did you even read any of the previous posts to your own which pre-empted your sentiments? You are clearly an adherent of business woo.
All the unsuccesful traders put in lots of time and effort as well, but they weren’t part of the group that, by pure statistical certainty there must be, had the good luck – as Damien P so clearly explained with his monkey coin flip example.
How have you managed to isolate these “deserving able” from the “undeserving unlucky”?
BJ said,
December 8, 2010 at 5:29 pm
general_jumbo: Could you refer me to the studies to back up your argument that their success is purely down to luck and not skill? Otherwise this just comes across as an unsubstantiated whinge.
So Goldman Sachs just happen to be lucky enough to employ incredibly lucky traders who by some fluke consistently earn over 50% of the company’s revenues every year? Do you really believe that?
QuietKnoll: Some simple maths shows you that if Soros was down 9 out of every 10 trades, he would not be a very rich man. Was it always pure luck that he wagered so much more on that one successful trade, or may we have to concede that the man does have some talent?
These traders you’re talking about are not comparable to the ‘mom and pop’ investors as it was put earlier, who are lose money trading for the simple reason that they don’t know what they’re doing, whereas there are a lot of traders out there who do.
leboeufsurletoit said,
December 8, 2010 at 11:51 pm
When asked for the secrets of his success, JP Getty said, “Rise early, work hard, strike oil”. Which, IMHO, says it all.
Gregory Goldmacher said,
December 10, 2010 at 5:57 pm
WRT to the investment debate shaping up, I’d like to toss in my $0.02 (how’s that for an investment!).
The causality here is not simple. Making each investment deal is like tossing a very heavy (20 pounds, say) and very expensive dart at a rotating dartboard. It takes effort, and you can affect the likelihood of success by putting in more work (exercise and practice; or, in investment, that’s studying the conditions and markets in an industry), but the outcome is also affected by chance. Each “hit” generates lots of money, which buys you more darts. In order to make a lot of money you have to be willing and able to throw a lot of darts.
Ceteris paribus, effort and skill do pay off. Luck plays a role too, of course. Success is also self-reinforcing, so people who get ahead can stay ahead (they can buy more darts to throw). It isn’t wrong of the people who succeed to attribute their success to work and skill, because those are a factor, but they do ignore the role of chance because of the natural human tendency to ascribe one’s successes to effort and one’s failures to the vagaries of circumstance.
jwm said,
December 11, 2010 at 2:16 pm
BJ:
“Some simple maths shows you that if Soros was down 9 out of every 10 trades, he would not be a very rich man. Was it always pure luck that he wagered so much more on that one successful trade, or may we have to concede that the man does have some talent?”
Actually, some rather simple maths shows that if the return on a gamble is more than ten times the investment, then even if you lose 9 times out of 10 youll still earn money.
Secondly, the stock markets continually grow (and hence the amazing stock market investors continually manage to make money) due to the continuous growth in the size of the worlds population, with its ever increasing demand for both essential and non essential goods. Therefore people who work on the stockmarket are “gambling” on the fact the world population will continue to grow and therefore sales on everything they “invest” in also continue to grow. We have a periodic crash because in their over enthusiasm they overestimate rate of growth. If they didnt do this they’d be fired because the growth in their investments would only mirror the growth of the economy and therefore not look very impressive to their managers.
The periodic crash brings stock prices down to the level which actually reflects the value of the economy they are investing in, rather than their inflated view of what they thought it would look like when they gambled 5 years before.
Ergo, no geniuses. Only fools playing with pretend money, with a guasian curve selecting where the money ends up during the upcurve that precedes the inevitable downcurve.
jwm said,
December 11, 2010 at 2:46 pm
For an example of this just look at Japan. 10 years of stagnation following a period of population stability. This has only reversed now that all shares reflect what they are actualy worth, as well as increased investment in exports and foreign economies where there is a continuing population growth.
Raggi said,
December 12, 2010 at 5:14 am
I don’t agree with QuietKnoll and Damien P. Of course everything is in the end luck related, don’t get me wrong. But you can have an affect on your luck.
Someone related this to gambling. When you invest a significant amount of money in a stock you just pick the best name, or close your eyes and pick a stock at random (this would imply that stock exchange is like roulette, witch would make it 50/50 +/- the fluctuation on the market…on average)
However ideally you don’t just invest in everything and anything. For example you wouldn’t invest in a new company today that was going to produce VHS movies. I know it isn’t this simple, but it’s still the same basic idea. In the bigger the better the odds on average, the more likely you’ll make any money out of it (kind of like poker)
rien said,
December 27, 2010 at 10:33 am
Put a monkey behind a PC for 10,000 years and he’ll leave a readable, consistent, yet not very infomative reply.
RoydeCosta said,
January 10, 2011 at 5:33 am
The imagination exercise described in this article shows how pseudo-sciences such as neuro-linguistic programming are promoted.
Context is all important though, and just like psychics who refer to unseen or subtle powers and give half promises of it working some of the time, neuro-linguistic programming authors set the rituals up first for the promise of power, and then for misattribution:
www.dailymotion.com/video/xdwl8h_characteristics-of-pseudoscience_tech
Btw, there are some interesting comments and accusations at the bottom of the video from irate scientologists and neuro-linguistic programming authors. Misattribution in action.
spaziomente said,
January 22, 2011 at 1:19 pm
Really an useful study. I’d like to share it in italian on www.spaziomente.wordpress.com.
funsizewo said,
January 24, 2011 at 9:07 am
Oh, shit it’s Monday
WilliamJay said,
February 15, 2011 at 4:44 pm
@BJ December 8, 2010 at 5:29 pm
“general_jumbo: Could you refer me to the studies to back up your argument that their success is purely down to luck and not skill? Otherwise this just comes across as an unsubstantiated whinge.”
You could try reading the Idiots Guide to Statistics. The good General is merely pointing out the fallacies that are prevalent in all gambling operations like Casinos and the Stock Market.
“So Goldman Sachs just happen to be lucky enough to employ incredibly lucky traders who by some fluke consistently earn over 50% of the company’s revenues every year? Do you really believe that?”
*sigh*, it’s like talking to a brick wall.
‘So Goldman Sachs just happen to be lucky enough’
ah the fallacy of faulty generalisation, the old ones are the best…
Five years ago someone could have made that argument with Lehman Brothers, or Citi Bank, or Bank of America, or Royal Bank of Scotland, etc. etc. etc. Today they would only dare make it with the name Goldman Sachs.
There are thousands of banks, new ones open every year, the many will ultimately fail, most will do reasonably, and a few will get very lucky ‘flip ten heads in a row’, and appear to be successful. (And as Soros points out magical-thinkers will then jump on the band wagon and the post hoc ergo propter hoc fallacy will take over, until of course it all goes horribly wrong)
But claiming ‘company X did this therefore this is a coherent business model’ is blatant selection bias. Other examples of this fallacy are: Feathered dinosaurs (birds) were very good at surviving the last major asteroid impact, therefore all dinosaurs were very good at surviving the last asteroid impact. Jenny took a homoeopathy pill and her cancer went into spontaneous remission, therefore every cancer patient who takes homoeopathy pills will go into spontaneous remission.
There may be many reasons why birds proved to be good at asteroid dodging, but a single case is not a credible guide to the current dinosaur population of the earth. There may be many reasons why a person’s cancer might go into remission, but the particular actions of a particular person are not scientific evidence. There may be many reasons why a particular business is successful one of which may be luck, and none of which need have anything to do with the viability of the industry it trades in.
“Whilst it’s fair to say successful traders will have some luck in their careers, they also put in huge amounts of time & effort analysing the markets.”
The average stock is now held for a grand total of 15 seconds. 15 seconds is long enough for the company’s investment potential to change dramatically? Long enough for people to research the solidity of their investment portfolio and change their mind? I think not. It is, however, long enough for the roulette wheel to spin. (There are of course the people who actually do analyse the markets and work out what things are worth – the Warren Buffets say – they however are an extreme exception)