Singer and colleagues (2004) investigated empathy for pain. The brain wasscanned when anticipating and watching a loved-one suffer a mild electric shock.There was an overlap between regions activated by expectancy of another person’spain and experiencing pain oneself, including the anterior cingulate cortex andthe insula. In a follow-up to this study, participants in an fMRI scanner watchedelectric shocks delivered to people who were considered either good or bad onthe basis of whether they had played fairly or unfairly in a game (Singer et al.,2006). While participants empathically activated their own pain regions whenwatching the “goodie” receive the electric shock, this response was attenuatedwhen they saw the “baddie” receiving the shock. In fact, male participants oftenactivated their ventral striatum (linked to better than expected rewards) whenwatching the baddie receive the shock—i.e. the exact opposite of simulationtheory. This brain activity correlated with their reported desire for revenge. Thissuggests that, although simulation may tend to operate automatically, it is notprotected from our higher order beliefs. Other research has shown that pain-relatedregions are activated differently when watching someone in pain depending onwhether one takes a self-centered or other-centered perspective and depending onone’s beliefs about whether the pain was necessary (Lamm et al., 2007). Thissuggests a significant amount of flexibility in mirroring that some simple versionsof simulation theory would not predict.