Next, the youths were interviewed. 'While I was out of the room, did you peek at any of the answers?' an experimenter asked. Eighty-four per cent of the peekers lied and said they hadn't peeked. Next they answered some questions about their understanding of truth and lying and the morality of dishonesty. Finally, all the participants were asked to promise to tell the truth in answer to the next question. This was a repeat of the question about whether they'd peeked at the answers. This time just 65 per cent lied - a statistically significant improvement.

Of course this first study doesn't show that the promise to tell the truth was the active ingredient in reducing lying - perhaps it was the discussion about morality or merely the act of being asked the same question twice. A second experiment with another forty-one 8- to 16-year-olds was identical to the first except the bit about promising to tell the truth was omitted. They still had the morality discussion and they were again asked twice whether they had peeked at the answers. Eighty-two per cent of peekers lied when first asked if they'd peeked. When asked again after the morality questions, 79 per cent still lied - no change in terms of statistical significance.

The lying youngsters in the first experiment who were asked to promise to tell the truth were eight times as likely to switch from lying to truth-telling than were the liars in the second experiment. 'When conducting forensic interviews with child and adolescent witnesses, police officers, social workers, and lawyers could use the honesty-promoting technique of promising to tell the truth,' the researchers said. 'In turn, the likelihood of obtaining truthful statements may increase.'

Evans AD, and Lee K (2010). Promising to tell the truth makes 8- to 16-year-olds more honest. Behavioral sciences and the law