Last time you told someone “I’ll call you,” did you mean it?
We all make promises in our daily interactions with others. On the one hand, promises such as “I’ll return your book next week” or “I won’t tell anyone” are not heavily binding, except maybe in a moral sense.

On the other hand, some of the promises we make bind us legally and financially. By saying “I do”, newlyweds promise to love and cherish each other no matter what happens for the rest of their lives; hardly anybody makes this promise intending to break it.

But imagine making a promise when in fact, you know you would benefit from not keeping it. Would you keep it anyway? Could we somehow tell in advance whether you’re going to keep it or break it?  And finally, could we predict your decision by looking at what happens in your brain?

All these questions are addressed in an exciting new study performed in Switzerland and led by Thomas Baumgartner and Urs Fischbacher. While their findings, published in "Neuron," are brand new and thus need to be confirmed by further research, they suggest that it may indeed be possible to detect whether a person is about to break a promise based on brain activity, well before the promise is actually broken.

The researchers ran a brain-scanning experiment in which pairs of participants played a well-established economic game involving trust. Player A, who was outside the MRI scanner, had to decide whether to keep or give away a certain amount of money -- say, $1 -- to Player B, who lay in the scanner.

If Player A decided to give the money to Player B, the amount would be increased five times (to $5). Once entrusted with the money, Player B could choose to either split it with Player A, so that each ended up with equal shares, or to keep it all.

Before Player A decided whether to hand over the money, player B made one of four promises: That if given the money, he would always, mostly, sometimes or never share it. The twist, however, was that once given the money, Player B could break his promise and keep the entire amount. Player A therefore faced a dilemma: to trust or not to trust the promise.

The main objective of this study was to illuminate how brain activity differs when promises are kept and when they are broken. Therefore, Player B’s brain activity was measured at three points during each game: When he made his promise; while he waited for player A to decide whether to trust him; and finally when he decided whether to keep his promise to share or not.

Interestingly, nearly all participants fell into one of two groups – about half were honest and consistently kept their promises, and the other half consistently broke them. The researchers compared the brain activity of the honest and dishonest players. They found that while breaking their promise, the dishonest players showed greater activity in regions of the brain known to be involved in generating and regulating emotional and cognitive conflict (the anterior cingulate cortex, parts of the dorsolateral prefrontal cortex, and the amygdala).

Fascinatingly, another network of regions in dishonest players’ brains (the anterior insula, anterior cingulate cortex and inferior frontal gyrus) showed increased activity while the players were making promises that they would later break. In other words, the dishonest players showed increased brain activity in several areas not only while breaking promises, but also at an earlier stage when their behavior was indistinguishable from that of honest players.

The researchers also examined whether any brain regions showed increased activation in the honest (compared to dishonest) participants, but found no such areas. They interpret this to mean that honesty may be a human ‘baseline’ – our brains might find it more effortful to be dishonest than honest.

This interpretation is rather optimistic about human nature, as it implies that we are hardwired to be honest, and that even those who consistently act dishonestly do not find it easy to do so. However, we must remember that failing to find a difference is not the same as proving that there is no difference. Honesty-related brain activity may be too subtle for such techniques to pick up.

In this study, each of the participants tended to be either honest or dishonest. Surprisingly, although we often think of honesty as a general personality trait, the researchers found no differences, on personality measures, between participants who broke promises and those who kept them.

Does this mean honesty is simply not part of one’s personality? Or perhaps we will never be able to capture honesty with a test? After all, being honest or dishonest involves a set of cognitive and social factors which may prove too complex to pin down.

The study provides an interesting insight into how deception can be investigated experimentally. Lately, research into the brain correlates of deception has been heating up. Most recent studies have investigated deception in the context of lie-detection, and they found increased brain activity in a number of regions, including the anterior cingulate cortex and dorsolateral prefrontal cortex.

However, such findings should be treated with caution, because the participants of most of these studies were instructed to tell a falsehood without having a choice about whether or not to do so. In the Baumgartner et al study, the researchers observed people doing what came naturally to them in a social interaction – and for about half, the experiment was enough to elicit deceptive behavior.

These findings, therefore, bear more relevance to real-life deception than those of most previous studies. They also show that deception can be successfully examined as a social act rather than in the context of lie-detection. This makes it possible to relate the findings to the wider spectrum of social behavior in which trust, cooperation and (unfortunately) deception are intimately linked.

The study opens up a host of questions for future research. For example, is dishonesty in economic decision-making the same as dishonesty in other situations, such as social, romantic or political interactions? Are dishonest people equally dishonest in different situations? And do similar brain mechanisms underlie all types of dishonest behavior?

So next time you say "I will call you" take a moment and ask yourself if you really mean it. You may be surprised to realize that you already know whether you do... or do not. This intention, we now know, is evident in your brain activity, so if you intend to break a promise, you might want to avoid making it in an MRI scanner.