Friends and strangers mess with your head

Separate teams probe the ways in which we decide who to be friends with, and who to avoid. Andrew Masterson reports.

Friendships tend to be based on similar age and interests – and neural architecture.
Friendships tend to be based on similar age and interests – and neural architecture.
Tara Moore / Getty Images

Friendships are driven by neural similarities, but the way we respond to strangers derives from a much less sophisticated approach, new research shows.

Two new papers investigating how individuals respond to people they know, and to people they don’t, have found the two processes to be markedly different.

A team led by psychologist Carolyn Parkinson from the University of California, Los Angeles, began with the uncontested observation that friendship groups tend to comprise people of broadly similar physical attributes, such as age and gender. Did similarity on external levels, the researchers wondered, coincide with similarities in the way friends perceive and respond to the world?

To find out, Parkinson and her colleagues selected 42 volunteers from a social cohort comprising 279 graduate students. The volunteers were all shown video clips that covered a wide range of topics and styles. As they watched, their brain activity was monitored, using functional magnetic resonance imaging (fMRI).

It’s perhaps no surprise that friends might have similar reactions to the same video footage – that they laugh at the same jokes, or are appalled by the same news content – but Parkinson’s team found these similarities ran very deep indeed.

“Neural responses when viewing audiovisual movies are exceptionally similar among friends,” the scientists found. Furthermore, they also found that the degree of neural response similarity decreases in line with the social distance between people in a friendship matrix.

The similarity in responses, the researchers write, drives relationships, so “people tend to be friends with individuals who see the world in a similar way.”

The fMRI results remained robust even after potentially confounding factors, such as differences in age or ethnicity, were taken into account. To further test their findings, Parkinson and colleagues ran their data through an algorithm trained to predict the closeness of relationship between any two people based solely on fMRI readings.

They found that the system correctly identified pairs of friends, or more distant associations, with a high degree of accuracy.

The research was published in the journal Nature Communications.

Meanwhile, elsewhere in the US, another group of researchers has been trying to understand the mechanism by which individuals decide to trust, or distrust, strangers.

A team led by Oriel FeldmanHall of the Department of Cognitive, Linguistic and Psychological Sciences at Brown University in Rhode Island, started from the premise that humans’ ability to learn through experience is useless when encountering a stranger. How then does anyone decide whether or not to trust at first meeting?

The answer, the researchers discovered, is through a process also found in many non-human species: stimulus generalisation. This is the same process that was exploited by nineteenth century Russian physiologist Ivan Pavlov, when, famously, he conditioned dogs to drool at the sound of a dinner bell, regardless of whether food was on offer.

In a paper published in the journal Proceedings of the National Academy of Sciences of the USA (PNAS), FeldmanHall and her colleagues report that strangers are assumed to be trustworthy if they resemble trustworthy people we already know.

To make this finding, the scientists first set up volunteers to play a “trust game” with three strangers. Each stranger had been schooled to act in a manner that was either completely trustworthy, partly trustworthy, or downright deceptive.

After each round of the game, the volunteers repeated it with three new strangers. Unknown to the volunteers, the new strangers were selected because they resembled the old ones – although the roles they played differed. The process was the repeated several times.

“Results reveal that subjects prefer to play with strangers who implicitly resemble the original player they previously learned was trustworthy and avoid playing with strangers resembling the untrustworthy player,” the researchers write.

And just as distances between people in social networks decrease in line with varying neural responses – as the University of California team discovered – decisions regarding trustworthiness also lose focus as physical appearance changes. The Rhode Island team found that as the resemblance between the original trusted character and each new stranger decreased, the decision whether to trust or distrust became less definite.

Contrib andrewmasterson.jpg?ixlib=rails 2.1
Andrew Masterson is news editor of Cosmos.
Latest Stories
MoreMore Articles