Monday 27 February 2017

Jumping to conclusions bias: How our minds actually work

I’ve been reading a book called Thinking: Fast and Slow by Dr Daniel Kahneman who is a Psychologist and won the Nobel Prize in Economics in 2002. His findings challenge the assumption that human rationality prevails in economic judgement.

In the book he writes at length of the two systems we use when making judgements or coming to conclusions. Simply put ‘System 1’ is our subconscious intuitive self that knows what is going on immediately and allows us to make very quick judgements. These can be judgements like the path of a ball through the air, or for an experienced driver, the minor change in lane position of a lorry you are driving past on the motorway that could mean the driver is not fully attentive on the road. The moment you register the movement you ‘know’ where the ball is headed or to give the lorry a little more space. System 2, in contrast, is our rational self the bit we think of as our internal monolog and the one that does the hard work to solve a difficult problem in a rational way. It’s not quite as simple as that but this explanation will suffice for this post.

Normally our System 2 receives the immediate answer from System 1 and simply rubber stamps it before moving on, but having said that if something is fully thought through our System 2 normally ‘wins’ over System 1 and we accept our System 2 judgement.

To put this into practice take a look at the picture below for a moment and ask yourself what is going on:

You immediately notice that this little chap is scared, you then see that he is all dressed up in his brand new school uniform and your System 1 serves up the narrative that he is worried about his first day at school, and you probably feel sorry for him and hope that it worked out ok in the end.

I’ve no idea if this is the right story because the picture was taken by ‘Andre2203’ and shared on a CC BY-SA 4.0 licence, but it’s the one we all jump to and it feels right to us. That’s our System 1 serving up a narrative to our System 2, which runs with it without modification. For all we know he might love school having been for the last 6 months and is actually scared of a big dog just out of shot.

Then there’s System 2, which is typified by answering the following calculation:

24 x 17 =

Our System 1 gives up on this immediately and hands it over to System 2 for analysis. We have all done quite a lot of maths in the past and our System 1 has some experience of this type of problem and can tell us that the answer is more than one or two hundred, but if 528 was suggested it couldn’t confirm or deny if that is the right answer. So, go on, try solving this question without a calculator and feel System 2 working. You may notice that 24 is almost 25 and that 25 x 17 is an easier place to start before taking the 17 off at the end. You can do 25 x 10 pretty easily so that just leaves the 25 x 7 left. And so on, and so on. Your short term memory quickly fills up with intermediate answers while trying to calculate the next part of the sum. It’s hard work, you probably looked away from the screen into space and your pupils will have dilated while you thought about it. At some point you will have either given up or found an answer and you pupils will have returned to their previous size. If you have not yet worked it out, the answer is 406.

So, let’s try an easier maths question where you will find the answer coming to mind effortlessly. The question goes like this:
A bat and a ball cost $1.10.
The bat costs $1 more than the ball.
How much does the ball cost?

The answer to this question came quickly and easily, System 1 is apparently good at this type of maths. But unfortunately you got the answer wrong. The answer you got was 10 cents, but if the ball cost 10c then the bat must cost $1.10 giving a total price of $1.20. So in fact the answer is not 10c, but 5c and now you know the answer provided by System 2 your System 1 is forced to agree.

I’ve tried this question on a few folks at work who are degree qualified professional engineers, and they, like me, all got it wrong the first time as well so don’t beat yourself up too much. This is a sort of mathematical illusion and works a bit like an optical illusion. But we should be aware that we can make this type of error and never be afraid of working something out long form. Don’t take the first answer you came up with for granted.

While we are here I’ll also point out that 24 x 17 is not 406 as I asserted above, it’s actually 408. If you spotted that error then well done, keep up the good work.

So to conclude, I hope that this has demonstrated that you and I have limits to our rationality. Even for something with a definite answer we often assume to ourselves that the first thing that comes to mind is the only answer and must be the right one and then don’t check for other alternatives. To think better we need to be aware of this ‘Jumping to conclusions bias’.

We also need to remember that even though we’ve thought it through carefully we still might not know the answer. We don’t actually know why the little boy in the picture looked scared so we don’t need to jump to a conclusion as to what is going on. It is acceptable to be ‘agnostic’ on that issue, and seek for better evidence like searching out and asking 'Andre2203' before concluding.

It’s also good to know that when pressed our System 2 can overrule the judgement our System 1 made; we can be more rational when we want to.

No comments:

Post a Comment