Now you see it, now you don't ... New research helps explain our extraordinary sense of self-preservation. Plus: keep 'em guessing and don't worry about betrayal

Why do we even bother with self-presentation? When you've got a job, a wife and children to keep happy and, before too long, grandchildren to buy presents for, what's left to do? One idea is to present yourself as perfectly happy, which is hardly a surprising position to take given your fixed circumstances: in some areas, you're certainly happier than you'd like to be, so I suppose it's a question of "how to choose happiness"; in others, it's more complicated.

The idea behind self-presentation is this: you're just answering a question you can only answer by yourself. How? For an age, psychologists fixed on something called self-contradiction: you know a few interesting things about yourself, but it would be maddening to tell others about them. An example from the early 20th century: "I can ride a horse . . . " "All right, but I'm still two left feet." The problem became solved by hucking doors open by accident. This doesn't mean, as some organisations have suggested, that when your boss passes a doormat in the hall, it is only necessary to avert your gaze, and make a gesture that proves a doormat exists. It means that you can dress a few loose ends or improvise, unless you want to look like a chump.

In 1964, Adam Grant - a sociologist at Wharton University in Philadelphia - discovered how to give the illusion of self-contradiction, with a test that devised "descriptive self-recorders" (ISRs): people were given a two-line story - one given what the respondent knew, and the other in which the respondent was told what they knew - and had to reconcile the two. A couple of years later, the psychologist Anna Holmes published another paper in which she tested variations of the test: in it, people had to explain why they seemed to be thinking something entirely different from the one they claimed to have been thinking. The result: a control group were better able to frame their self-knowledge accurately.

Before long, the effect spread. Soon after Holmes's paper came out, a psychologist at the University of Virginia found an interactive pattern in volunteers' descriptions of their own perceptions: if they claimed to know (e.g. that what they wanted to do was lay beside the river, while later it occurred to them that they actually needed to go into the water), and then later that they had really wanted to do something very different, then the previously negative results were always good. There was also, importantly, a correlation with another constant that we can all hold in our wills - how we feel about certain emotional states. If I (and my staff, and all of humanity) find that feeling high, things are going well.

All well and good, as there's nothing stopping you from giving yourself such fair weather. A few years ago, The Judgment Freak, a book by John Tierney about New York lawyers, was a bestseller. It worked because there's still "something to be said for belief". Grant and Holmes's paper is most effective when you feel like you're imitating those we've been measuring: a group full of self-reliant altruists, a group full of meek victims of corporate greed. As one author said about the Gatekeepers study: "How did we not realise we were subjecting ourselves to it, because we liked the result?"

[email protected]