Our personality is the quality that forms our character – it’s the driver behind our interactions, behaviour, and emotions.
As human beings, we make use of personality judgements on a daily basis to make decisions about our personal and professional lives; decisions like who to befriend, marry, or employ, and the more accurate the judgement, the better the decision.
It’s said that we use social cognition (the way we process, store, and apply information about other people) to understand other people’s personality. But research done by the University of Cambridge and Stanford University show that computers can also make accurate judgments of a person’s personality and it’s all thanks to what we ‘like’ on Facebook.
As Facebook users, we make use of the ‘Like’ function to express a positive association with subjects (online or offline) such as music, books, sport, people, products, and websites. Plain and simple – it’s the stuff we like.
The findings, published in the Proceedings of the National Academy of Sciences, show that a computer only has to analyse about 10 Facebook likes to make a more accurate personality judgement than a one of our work colleagues can. The same goes for a friend, a family member, and a spouse where the computer only requires 70, 150, and 300 likes, respectively, to outperform their personality judgements of us.
The study compared the accuracy of computer personality judgements to that of human judgements. The researchers used a sample of 86 220 volunteers who each completed a 100-item self-reported personality questionnaire measuring traits of openness, conscientiousness, extraversion, agreeableness, and neuroticism (the big five traits).
Computers were given access to 90% of these volunteers’ Facebook likes while a friend or family member of each volunteer had to complete a 10-item questionnaire about that particular volunteer in an effort to determine their personality.
The researchers explained that likes can be used to diagnose one’s personality because it represents activities, attitudes, and preferences aligned with the big five traits. If, for example, a person likes meditation or TED talks on Facebook, it signifies openness while extroversion could be indicated by liking subjects such partying or dancing.
By comparing the computer judgements to that of the friends or family, it was found that with enough ‘Likes’, a computer could make a better personality judgement than a person’s friends or family could.
With this study, the researchers noted that computers may have a couple of advantages over human beings when it comes to personality judgement. Accurately judging a person’s personality depends on the availability and the amount of relevant behavioural information, along with the judge’s ability to detect and use it correctly.
Computers are able to easily store all this information and because human judgements tend to include certain motivational biases, computer judgement might be more accurate.
However, humans have the advantage of being flexible and able to capture many subconscious cues unavailable to computers which mean we might be better at describing traits that require subtle cognition which aren’t available in digital behaviour.
Using computers to make personality assessments can affect society in many ways. In future, employers might use it to match candidates with jobs suited to their personality. People might even turn to computers to help them make important decisions, like what career path to follow or who they should marry.
The possibility does exist that such data-driven decisions might improve people’s lives.