Your Computer Knows You Better Than Your Friends and Family

The average person’s computer and online social media accounts contain a vast array of information about their interests and background. In fact, a group of researchers from Stanford University and the University of Cambridge Massachusetts say a person’s Facebook page paints a better picture of their personality than assessments conducted by friends and family members.

Researchers wanted to see how well a computer could select personality traits that are normally predicted by humans. In other words, they set up a study in which computer systems assessed the personality traits of users – traits that are considered to be important social-cognitive activity. Researchers concluded that computers performed a better job at assessing users’ personality than humans.

How did researchers come to this conclusion? The study involved the collection of personality traits (known as markers) from 86,220 participants, all of whom were asked to complete a 100-question survey on the Android/iOS app “myPersonality.” This app is designed to measure the following five personality markers: extraversion, openness, conscientiousness, neuroticsm, and agreeableness.

Participants’ friends and family members were later asked to complete a shorter 10-question survey about the participant, while the computer system simultaneously completed the survey, basing its answers strictly on the participants’ Facebook likes instead. Surprisingly, the computer systems outperformed the participants’ friends and family members when gauging their personality markers.

The report reveals that computer systems outperformed participants’ work colleagues using the information from just 10 Facebook likes. And after viewing 70 Facebook likes, the computer outscored participants’ friends and roommates. When it based its answers on 150 Facebook likes, however, the computer system scored higher than participants’ friends and family members.

Researchers say a person’s Facebook likes tell a lot about themselves. For instance, a person who likes meditation is more likely to be artistic and liberal. The only criteria in which humans outperformed their computer counterpart was life satisfaction, which is something that you apparently can’t tell from looking at a person’s Facebook likes.

“Computers’ judgments of people’s personalities based on their digital footprints are more accurate and valid than judgments made by their close others or acquaintances (friends, family, spouse, colleagues, etc.),” wrote the study’s researchers. “Our findings highlight that people’s personalities can be predicted automatically and without involving human social-cognitive skills.”

This study was published in the journal Proceedings of the National Academy of Science (PNAS)