Images of Altruism

Three different people have e-mailed me this study, which seems to be making the rounds in the news.

The article, published in an upcomming issue of Nature Neuroscience, looked at fMRI images of subjects' brains as they played a game with a computer. They also asked subjects to complete a range of measures, including altruism, empathy and other personality questionnaires. They found that activity in the posterior superior temporal cortex (pSTC) during the game was associated with self-reported altruism but not with any of the personality measures.

They point out that the pSTC has previously been linked to understanding of intentionality. For instance, when subjects see geometric objects moving with seeming purpose across a screen, the pSTC becomes active, but not when the objects move at random. The authors suggest that altruism requires an understanding of agency - what I would call, although the authors do not, Theory of Mind. Because pSTC activity was not linked to the personality measures such as empathic tendency, they suggest that the pSTC acts independently of empathy and other personality measures.



I'm not surprised by the result. Why would you help someone unless you appreciate that they're an indepedent being, capable of acting with purpose and therefore also capable of suffering? Very young children don't grasp this concept. In a common experiment, children are shown a girl hiding a doll in one of two chests. While she's gone, another person comes in and switches the doll to the other chest. When the girl comes back, the children are asked which chest she should look in. Younger children answer with the chest the doll has been moved to. They can't separate what they know from what the girl knows. They don't quite understand that other people have independent minds like their own. Interestingly, the age when children start to answer "the chest she put it in, she thinks it's still there", is around the same age they start engaging in genuine helping behavior.

The authors of this study found no relationship between understanding - or rather, appreciating - the agency of others, at least as measured by pSTC activation, and empathy. They used the Interpersonal Reactivity Index, the same questionnaire I'm using in my Div III. Looking at their supplemental data you can see that there was a correlation between the empathic concern subscale and pSTC activation. There was a correlation between the perspective-taking subscale of the IRI and OFC activation as well. However, subsequent narrowing down of activation areas into "regions of interest" made this correlation not significant - I don't understand the reasoning for that narrowing-down, as I'm not experienced with MRI studies, but it does seem to me that a relationship between pSTC activation and empathy can't be entirely ruled out.

My other issue with the study is that it's focused around the idea of understanding the agency of others. The game they play is focused on that, as opposed to on straight-out altruistic behavior (subjects played trials for their own gain and for charity, but this was only taken into account after the area of interest was narrowed to the pSTC and the OFC). Basically, what they've established is that the pSTC has some sort of influence on altruism. That's great, but there are a lot of factors that are important for altruism. Sociability, for instance. Because most opportunities for helping behavior require interpersonal interaction of some sort, shy people tend to be less altruistic. My own Div III looks at another factor, tendency to feel personal distress. Other researchers have looked at feelings of social responsibility. All of these are correlated with altruism, and all of them may involve activation of areas in the brain measurable by fMRI. None of these are measured in the study. Which is not to say that this study should have looked at them. It's important to take baby steps. But... I'm not sure the sweeping claims that some articles are making (shame on you, BBC!) are justified.