What Schachter and Singer showed is that we interpret the same physiological state not merely as different versions of the same emotion (e.g. being envious of different things) but as examples of different emotions entirely (anger versus elation). And this is perhaps surprising because it suggests that our ‘read-out’ of our physiological state –

Building intelligent machines would be off to a flying start if it happened that we could just ask people what they knew – and write this knowledge directly into a computer database.

If we are asked to choose an option, we mostly focus on reasons for choosing one thing or another: and these reasons will tend to be positive reasons in favour of one option or the other. The extreme option has the most powerful positive reasons (e.g. a very close relationship with the child), so it wins out. If, on the other hand, we are asked to

Our thoughts are not shadows of an alternative inner reality, to be charted and discovered; they are fictions of our own devising, created moment by moment.

But the problem with our everyday view of our minds is far deeper: no one, at any point in human history, has ever been guided by inner beliefs or desires, any more than any human being has been possessed by evil spirits or watched over by a guardian angel.

In short, common-sense psychology sees our thought and behaviour as rooted in reasoning, but a lot of human intelligence seems to be a matter of finding complex patterns.

An improvising mind, unmoored from stable beliefs and desires, might seem to be a recipe for mental chaos. I shall argue that the opposite is true: the very task of our improvising mind is to make our thoughts and behaviour as coherent as possible – to stay ‘in character’ as well as we are able.

Yet the contents of our hidden depths seem to remain perpetually elusive. Freudian psychoanalysts can speculate about our hidden fears and desires; psychologists and neuroscientists can attempt to draw subtle and highly indirect conclusions from actions, heart-rate, skin conductance, pupil dilation and the rate of blood flow in the brain. But no

The danger is that one moment’s speculative thought (‘I don’t love Alys’, ‘I’m a hopeless failure’, ‘the world is terrifying’) becomes the next moment’s incontrovertible proof – the very thought is taken as its own justification.