Sunday, 16 February 2014

Fundamental Attribution Error

I have been reading a little, and thinking a little, about Fundamental Attribution Error (see, for example, here).  It is fascinating: the basic premise is that we tend to attribute more direct intentionality to human behaviour than is in fact appropriate.  That is, someone does something, and we think that tells us something about their character, and attribute characteristics to them. So, for example, someone cuts us up in the traffic queue, and we assume firstly that the act was deliberate (ie we exclude the likelihood of simple mistakes) and secondly that they are selfish (ie we exclude the possibility of other influences on their behaviour, such as being late for a train).

All this was ringing a faint bell, and then the penny dropped:
If any human acts may loosely be called causeless, they are the minor acts of a healthy man; whistling as he walks; slashing the grass with a stick; kicking his heels or rubbing his hands. It is the happy man who does the useless things; the sick man is not strong enough to be idle. It is exactly such careless and causeless actions that the madman could never understand; for the madman (like the determinist) generally sees too much cause in everything. The madman would read a conspiratorial significance into those empty activities. He would think that the lopping of the grass was an attack on private property
Chesterton, of course, in the second chapter of Orthodoxy.  

And of course, once we have attributed characteristics  to somebody, our old friend confirmation bias means that they tend to stay attributed.  Once someone has decided (to take an example purely at random, you understand) that Ben Trovato is bad, he can continue to collect evidence to that effect, and (in particular) to interpret my sayings and doings through a hermeneutic of suspicion, and prove (to his own satisfaction, at least) that he is right.

Clearly, that is very visible to me, as I have a different bias, in my own favour. But the more interesting part is really this: the realisation that if it is clear to me that others do this, then the likelihood is that I do it too.

And once one starts reflecting on that, the potential for learning and understanding is intriguing, to say the least.

And it sheds an awful lot of light on Twitter...

No comments: