By Emma Young
We make all kinds of snap decisions about a person based on their facial appearance. How trustworthy we think they are is one of the most important, as it can have many social and financial consequences, from influencing our decisions about whether to lend someone money to which Airbnb property to book.
However, as the authors of a new study, published in the British Journal of Psychology, note, “Although facial impressions of trustworthiness are formed automatically, they are not especially accurate predictors of trustworthy behaviour.” People who are less susceptible to forming these impressions could, then, be at an advantage. And, as Jasmine Hooper at the University of Western of Australia and colleagues now report, men with high levels of autistic traits fall into this category.
Earlier work (involving two authors of the new study) already hinted that autism might have a bearing on whether people act on first impressions in a typical way. The researchers found that when they prompted boys diagnosed with autism to judge trustworthiness from photographs, they formed the same kinds of impressions as non-autistic boys, but unlike them, they did not use these judgments in their decision-making in an economic trust game.
To explore whether higher levels of autistic-like traits, such as increased attention to detail, might also affect the use of first impressions even among the general population, the researchers recruited 46 men without autism, who completed the Autism-Spectrum Quotient (a symptom-based scale which is typically used by clinicians to diagnose autism) and a facial trustworthiness impressions task. The researchers found that, as they predicted, when prompted to make trustworthiness judgements from faces, people with high or low levels of autistic traits made similar judgements. But the next phase of the study showed some important differences between the two groups.
All the participants also took part in an adult version of the economic trust game used in the earlier research with boys, in which they interacted with virtual “partners”, who had been independently rated as looking either trustworthy or untrustworthy. Over a series of trials, these partners were consistently fair or selfish, and crucially this behaviour pattern was not related to their facial appearance. Therefore the best strategy for predicting future behaviour – and making money, because participants could keep their winnings – was for the participants to ignore appearances and focus solely on their partner’s behaviour in previous money allocation trials.
Again as the researchers predicted, the men with relatively low levels of autistic traits were influenced by how trustworthy or not the partners looked. The men who scored relatively high on the AQ scale were influenced too, but to a significantly lesser extent – they were more influenced by their partners’ actual behaviour.
Only between 1 and 1.5 per cent of people in the UK are thought to meet the criteria for a diagnosis of autism spectrum disorder. However, many other people do not meet the diagnostic criteria for autism but do have relatively high levels of autistic traits – there may therefore be a sizeable subset of people who are able to form trustworthiness judgements based on people’s faces when prompted, but who don’t do this automatically, and also aren’t so inclined to use these judgments in their social decisions (and so are less swayed by potentially misleading facial impressions than the rest of the population).
As the researchers note, “If facial impressions are not especially accurate predictions of actual trustworthiness, failure to use them to guide trust decisions may actually represent highly rational social behaviour.”