Peer pressure
I try not to give money to subway panhandlers: They often seem not to be genuine, or I find myself with no spare cash. But I would be lying if I didn’t admit that I sometimes cave in to peer pressure, not wanting to be the lone curmudgeon in a sea of caring commuters.
In this case, my motivation for donating is not really generosity, but concern for my social reputation — a well-documented phenomenon. Individuals with autism, however, are no more likely to donate money when being observed than when alone, reports a study published 18 October in Proceedings of the National Academy of Sciences1.
In the study, researchers gave $45 to 10 individuals with autism and 11 controls. The participants then went into another room and either accepted or rejected a series of computerized requests to give some of this money to the United Nations Children’s Fund.
In these scenarios, the amount of money the participant was asked to spend did not always match the amount the charity received. For example, one scenario asked the participant to donate $16, but the charity would receive only $4; in another, participants gave $4, but the charity would receive $16.
The computer program asked the participants to agree or disagree to 50 different requests, and told them they would be required to follow through with just one, chosen randomly at the end of the day.
Individuals with autism and controls all donated money similarly and somewhat predictably, although the individuals with autism gave slightly less. For example, both groups were unlikely to accept requests in which they donated more than the charity received.
The researchers also varied the test with a different scenario, this time telling the participants that because of a computer error, a third person would observe and record their choices.
In this situation, controls accepted more of the donation choices and responded faster, perhaps wanting to seem more willing to give. Being watched also had the biggest influence on those controls who had donated the least to begin with.
Participants with autism showed no such change in their behavior and, if anything, donated a little less than they had before.
To test whether this difference reflected concern for social reputation and not just awareness of being watched, the researchers repeated the observer setup with a straightforward attention test. In this test, participants were asked to click a button only when they saw an ‘X’ on the computer screen. Both groups responded more quickly when watched.
Studies like this one aim to parse out the brain pathways that underlie the social deficits seen in people with the disorder. For example, individuals with autism might find social acceptance to be less rewarding than controls do, the researchers say, a speculation backed up by brain imaging experiments.
Those with autism might also have trouble understanding how others see them, which is in line with deficits in theory of mind — the ability to understand others’ thoughts and beliefs.
Either way, the results suggest one more intriguing piece in the puzzle of social behavior in autism. They also explain why it often takes just one subway passenger reaching into a wallet to inspire generosity in a carful of riders.
References:
1: Izuma K. et al. Proc. Natl. Acad. Sci. USA 108, 17302-17307 (2011) PubMed
Comments
Sorry to always be making negative comments about papers, but this one really doesn't stand up to scrutiny.
There's a scatterplot in the paper, which shows the actual scores (well the difference between the observed and the not-observed condition) for each participant.
Only 4 of the 11 control participants showed an observer effect. In other words, the absence of an observer effect (deemed to be a sign of social impairment in autism) is also reported in the overwhelming majority of controls.
Jon- Nothing wrong with a careful critique of the results.
I actually count 6 controls with at least a slight observer effect. But, you are correct, the numbers are very low. Thought it was a potentially interesting and provocative finding, though, which is why we wrote it up as a blog.
I should have been clearer. One of the controls and three of the autism group show a *negative* observer effect of at least 2 points, which suggests that effects of +/- 2 points is not a meaningful effect.
Using those criteria, 4 controls showed a positive effect, 7 controls and 8 people with autism showed no effect, and two people with autism showed a negative effect.
I don't think we can conclude from this that people with autism have no sensitivity to peer pressure (or "empathy" as the authors argue). If we do, then we have to say that most non-autistic people lack this sensitivity too!!