regarded social integration as a type of reward (Baumeister and Leary, 1995).

Dr. Zaki noted that reinforcement learning has two essential elements: (1) prediction of reward, and (2) adaptation. While the brain responds to rewards, it eventually learns to predict when such rewards occur by identifying social and other temporal cues preceding the presentation of a reward (O’Doherty et al., 2003). A well-known example of this is Pavlov’s experiments in which dogs identified the ringing of a bell with a reward of a treat. What is noteworthy in this, Dr. Zaki points out, is that the brain eventually stops responding to the reward itself and instead responds to the predictor of the reward. This response occurs in the nucleus accumbens. In addition to predicting rewards, people adapt behavior in order to actively gain rewards. Often this adaptation occurs in ways in which people do not necessarily understand.

Given this, Dr. Zaki posited that one predictor of the reward of social integration is consensus with other people. This indicates that an individual will not only respond to agreement with other people, but will also actively seek such conformity, even to the point of changing their own behavior. Dr. Zaki demonstrated through research that this seems to be the case, both psychologically and neurologically. In one such study, participants were asked to rate attractiveness of various faces. They were then told the ratings their peers gave. Upon being asked to rate the faces again, the responses of the participants changed, at a statistically significant level, so as to conform to the opinions of their peers. He also noted that this change occurred not just at a superficial level, but also at a deeper level in which participants truly believed in their new opinions, as observed by viewing whether the structures of the brain responding to facial attractiveness were activated (Zaki et al., 2011). Dr. Zaki speculated that, given the power of this mechanism, reinforcement learning could indicate that those in situations of high violence might be internalizing their attempts at conformity. Over time, this may lead to intractability. He also noted, however, that research indicates this mechanism is very generalizable, and it would be useful to consider how it could be used to promote the spread of prosocial behavior.


Psychological and physiological processes within individuals have the potential to shed light on the mechanism of the contagion of violence. Speakers also noted the importance of such processes interacting with the context or environment in which stimuli, observation, and conformity occur. Other speakers considered the importance of that context or environment itself, as well as the interplay of multiple factors. In particular, the role of group dynamics, alluded to previously in terms of collective aggression,

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement