On receiving a paypacket, how good a man feels depends on how much his colleague earns in comparison, scientists say.
Brain scans show we measure our success by others' earnings
Scans reveal that being paid more than a co-worker stimulates the "reward centre" in the male brain.
Traditional economic theory assumes the only important factor is the absolute size of the reward.
But researchers in the journal Science have shown the relative size of one's earnings play a major role.
In the study, 38 pairs of male volunteers were asked to perform the same simple task simultaneously, and promised payment for success.
Both "players" were asked to estimate the number of dots appearing on a screen. Providing the right answer earned a real financial reward between 30 (£22) and 120 (£86) euros. Each of the participants was told how their partners had performed and how much they were paid.
Using magnetic resonance tomographs, the researchers examined the volunteers' blood circulation throughout the activities. High blood flow indicated that the nerve cells in the respective part of the brain were particularly active.
Brain scans showing high activation (top) and low activation (bottom)
Neuroscientist Dr Bernd Weber explains: "One area in particular, the ventral striatum, is the region where part of what we call the 'reward system' is located. In this area, we observed an activation when the player completed his task correctly."
A wrong answer, and no payment, resulted in a reduction in blood flow to the "reward region". But the area "lit up" when volunteers earned money, and interestingly showed far more activity if a player received more than his partner.
This indicated that stimulation of the reward centre was not merely linked to individual success, but to the success of others.
While behavioural experiments have suggested relative rewards may play a role in economic motivation, economist Professor Dr Armin Falk, co-author of the paper, said: "It is the first time this hypothesis has been challenged using such an experimental approach."
The professor emphasised to BBC News, that unlike behavioural experiments, brain scans had "no cognitive filter; we were monitoring immediate brain reaction".