Published: 30/06/2005, Volume II5, No. 5962 Page 36

Trust and its exploitation are at the heart of office politics and among the strategies of ambitious employees in any organisation.

Imagine someone who had evolved the ability to tell lies and not give away what they are up to because they had perfect control over body language and vocal tremors. They should be amazingly successful, given their ability to exploit the rest of us, and more so than someone who always gave themselves away the moment they tried to slip a fast one past the boss.

This leads to a quandary at the heart of evolutionary psychology: why has perfect deception not evolved? Most of us have a go at telling lies and are sometimes successful, but often we lack the confidence in our ability to spin our way out of trouble and stick to the straight and narrow probably more than we would like to.

Paul Andrews, an evolutionary psychologist at the University of New Mexico, has come up with an intriguing theory about our tendency to lie imperfectly. He mathematically modelled the best strategy in trust games. These are experimental scenarios where participants are asked to trust others by giving them benefits, uncertain as to whether these will be reciprocated or shared.

The problem with perfect deception is that no-one would want to enter a personal or business relationship with you if they felt totally unable to 'read' you. They wouldn't trust you. Instead, because we know that deceit is an inevitable part of human affairs, it is our confidence in spotting deceit in others that leads us to trust.

The truly perfect strategy is to appear to be an imperfect liar; to allow another to believe they can read you by allowing deceit to be detected at times, retaining the ability to put one over undetected when it really counts.

It is in our interests to 'leak' our true intentions from time to time because that makes others believe they are in control of the trust in the relationship.

Could this be the real reason why no perfect liar has yet been recruited into a study? It is not in the interests of a perfect liar to be detected.

Dr Raj Persaud is consultant psychiatrist at South London and Maudsley trust and Gresham professor for public understanding of psychiatry.