New Stanford research finds computers are better judges of personality than friends and family
Stanford researchers have found that computers can judge personality traits more accurately than one's friends and colleagues. In fact, artificial intelligence can draw inferences about a person as accurately as a spouse, according to Stanford postdoctoral fellow Michal Kosinski.
New research shows that a computer's analysis of data can better judge a person's psychological traits than family and friends.
Computers can judge personality traits far more precisely than ever believed, according to newly published research.
In fact, they might do so better than one's friends and colleagues. The study, published Jan. 12 and conducted jointly by researchers at Stanford University and the University of Cambridge, compares the ability of computers and people to make accurate judgments about our personalities. People's judgments were based on their familiarity with the judged individual, while the computer used digital signals – Facebook "likes."
The researchers were Michal Kosinski, co-lead author and a postdoctoral fellow at Stanford's Department of Computer Science; Wu Youyou, co-lead author and a doctoral student at the University of Cambridge; and David Stillwell, a researcher at the University of Cambridge.
According to Kosinski, the findings reveal that by mining a person's Facebook "likes," a computer was able to predict a person's personality more accurately than most of their friends and family. Only a person's spouse came close to matching the computer's results.
The computer predictions were based on which articles, videos, artists and other items the person had liked on Facebook. The idea was to see how closely a computer prediction could match the subject's own scores on the five most basic personality dimensions: openness, conscientiousness, extraversion, agreeableness and neuroticism.
The researchers noted, "This is an emphatic demonstration of the ability of a person's psychological traits to be discovered by an analysis of data, not requiring any person-to-person interaction. It shows that machines can get to know us better than we'd previously thought, a crucial step in interactions between people and computers."
Kosinski, a computational social scientist, pointed out that "the findings also suggest that in the future, computers could be able to infer our psychological traits and react accordingly, leading to the emergence of emotionally intelligent and socially skilled machines."
"In this context," he added, "the human-computer interactions depicted in science fiction films such as Her seem not to be beyond our reach."
He said the research advances previous work from the University of Cambridge in 2013 that showed that a variety of psychological and demographic characteristics could be "predicted with startling accuracy" through Facebook likes.
The study's methodology
In the new study, researchers collected personality self-ratings of 86,220 volunteers using a standard, 100-item long personality questionnaire. Human judges, including Facebook friends and family members, expressed their judgment of a subject's personality using a 10-item questionnaire. Computer-based personality judgments, based on their Facebook likes, were obtained for the participants.
The results showed that a computer could more accurately predict the subject's personality than a work colleague by analyzing just 10 likes; more than a friend or a roommate with 70; a family member with 150; and a spouse with 300 likes.
"Given that an average Facebook user has about 227 likes (and this number is growing steadily), artificial intelligence has a potential to know us better than our closest companions do," wrote Kosinski and his colleagues.
Why are machines better in judging personality than human beings?
Kosinski said that computers have a couple of key advantages over human beings in the area of personality analysis. Above all, they can retain and access large quantities of information, and analyze all this data through algorithms.
This provides the accuracy that the human mind has a hard time achieving due to a human tendency to give too much weight to one or two examples or to lapse into non-rational ways of thinking, the researchers wrote.
Nevertheless, the authors concede that the detection of some personality traits might be best left to human beings, such as "those (traits) without digital footprints and those depending on subtle cognition."
Wu, co-lead author of the study, explains that the plot behind a movie like Her (released in 2013) becomes increasingly realistic. The film involves a man who strikes up a relationship with an advanced computer operating system that promises to be an intuitive entity in its own right.
"The ability to accurately assess psychological traits and states, using digital footprints of behavior, occupies an important milestone on the path toward more social human-computer interactions," said Wu.
Such data-driven decisions could improve people's lives, the researchers said. For example, recruiters could better match candidates with jobs based on their personality, and companies could better match products and services with consumers' personalities.
"The ability to judge personality is an essential component of social living – from day-to-day decisions to long-term plans such as whom to marry, trust, hire or elect as president," said Stillwell.
The researchers acknowledge that this type of research may conjure up privacy concerns about online data mining and tracking the activities of users.
"A future with our habits being an open book may seem dystopian to those who worry about privacy," they wrote.
Kosinski said, "We hope that consumers, technology developers and policymakers will tackle those challenges by supporting privacy-protecting laws and technologies, and giving the users full control over their digital footprints."
In July, Kosinski will begin a new appointment as an assistant professor at Stanford Graduate School of Business.
Michal Kosinski, Stanford Computer Science: (650) 739-5679, firstname.lastname@example.org
Clifton B. Parker, Stanford News Service: (650) 725-0224, email@example.com
Dan Stober, Stanford News Service: (650) 721-6965, firstname.lastname@example.org