For many years, evolutionary biologists, behaviorists, economists and political scientists have attempted to understand why cooperation exists between human beings, even though that cooperation may not result in a direct or immediate reward. This unselfish behavior called "altruism" is almost uniquely a human trait.
Up until now, almost all brain imaging experiments that have studied the social brain have done so by exposing subjects to static 2-D images inside the scanner. "This study represents an attempt to learn about the social brain by scanning people as they are engaged in a true social interaction," said James K. Rilling, Ph.D., principal investigator in the Emory study, who is currently serving a postdoctoral fellowship at Princeton University. In the Fall of 2003, Dr. Rilling will return to Emory as a faculty member with a joint appointment in the Center for Behavioral Neuroscience (CBN) at Emory University School of Medicine and the Emory University Department of Anthroplogy.
In two separate experiments, the researchers used fMRI to scan the brains of 36 women while they played the "Prisoner's Dilemma Game," a decades-old model for cooperation based on reciprocal altruism. Two players independently chose to either cooperate with each other or not (defect), and each was awarded a sum of money that depended upon the interaction of both players' choices in that round.
In the first experiment, 19 subjects were scanned in four game sessions designed to observe neural function during cooperation and non-cooperation during both human interactions (social) and interactions with a computer (non-social). The results of the first experiment revealed different patterns of neural activation depending on whether the playing partner was identified as a human or a computer. In the second experiment, 17 subjects were scanned during three game sessions, focusing specifically on human interaction.
Mutual cooperation was the most common outcome in games played with presumed human partners in both experiments, even though a player was maximally rewarded for defecting when the other player cooperated. During the mutually cooperative social interactions, activation was noted in those areas of the brain that are linked to reward processing: the nucleus accumbens, the caudate nucleus, ventromedial frontal/orbitofrontal cortex and rostral anterior cingulate cortex.
"Our study shows, for the first time, that social cooperation is intrinsically rewarding to the human brain, even in the face of pressures to the contrary, " said Gregory S. Berns, M.D., Ph.D., co-investigator and associate professor of psychiatry in the Emory University School of Medicine Department of Psychiatry and Behavioral Sciences and member of the CBN. "It suggests that the altruistic drive to cooperate is biologically embedded-- either genetically programmed or acquired through socialization during childhood and adolescence."
"Reciprocal altruism activates a reward circuit, and this activation may often be sufficiently reinforcing to override subsequent temptations to accept but not reciprocate altruism. This may be what motivates us to persist with cooperative social interactions and reap the benefits of sustained mutual cooperation," said Dr. Rilling.
"The combination of game behavior and functional brain imaging also provides a unique paradigm to explore the neural basis of social behavioral disorders such as autism, drug addiction and sociopathy, that are characterized by deficits in social reciprocity, impulse regulation, or social reward processing," adds Clint Kilts, Ph.D., co-investigator and associate professor of psychiatry at Emory. "It defines the most complex form of the human genesis of a social bond. It may help us define why wars are fought and loves are lost."
The study was sponsored by the Markey Center for Neurological Sciences Fellowship, National Institute on Drug Abuse (NIDA), National Institutes of Mental Health (NIMH) and National Alliance for Research on Schizophrenia and Depression (NARSAD). Other Emory researchers involved in the study were David A. Gutman, Thorsten R. Zeh, and Giuseppe Pagnoni, Ph.D.
Journal
Neuron