Computer-based personality judgments are more accurate than those made by humans
By Youyou Wu, Michal Kosinski, David J. Stillwell. 2015
959 citations recorded by the source corpus.
Significance This study compares the accuracy of personality judgment—a ubiquitous and important social-cognitive activity—between computer models and humans. Using several criteria, we show that 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.). Our findings highlight that people’s personalities can be predicted automatically and without involving human social-cognitive skills. Judging others’ personalities is an essential skill in successful social living, as personality is a key driver behind people’s interactions, behaviors, and emotions. Although accurate personality judgments stem from social-cognitive skills, developments in machine learning show that computer models can also make valid judgments. This study compares the accuracy of human and computer-based personality judgments, using a sample of 86,220 volunteers who completed a 100-item personality questionnaire. We show that (i) computer predictions based on a generic digital footprint (Facebook Likes) are more accurate (r = 0.56) than those made by the participants’ Facebook friends using a personality questionnaire (r = 0.49); (ii) computer models show higher interjudge agreement; and (iii) computer personality judgments have higher external validity when predicting life outcomes such as substance use, political attitudes, and physical health; for some outcomes, they even outperform the self-rated personality scores. Computers outpacing humans in personality judgment presents significant opportunities and challenges in the areas of psychological assessment, marketing, and privacy.
Subfields: Clinical Psychology, Experimental and Cognitive Psychology, AI Ethics, Fairness & Society, Cognitive Science & Psychology
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Clinical Psychology
- Lyle Ungar
- Davide Marengo
- Michele Settanni
- Danny Azucar
- Daniel Preotiuc-Pietro
- Michal Kosinski
- David J. Stillwell
- Stephan Lewandowsky
In Experimental and Cognitive Psychology
- Iason Gabriel
- Daniel Preotiuc-Pietro
- Lyle Ungar
- Paolo Rosso
- Benno Stein
- Albert Gatt
- Cornelius J. Konig
- Ponnurangam Kumaraguru
In AI Ethics, Fairness & Society
- William S. Isaac
- Iason Gabriel
- Lyle Ungar
- Iyad Rahwan
- Borja Balle
- Jacob William Crandall
- Lucie Flekova
- Marise Ph. Born
In Cognitive Science & Psychology
- Iason Gabriel
- Daniel Preotiuc-Pietro
- Lyle Ungar
- Albert Gatt
- Iyad Rahwan
- Cornelius J. Konig
- Azim Shariff
- Emiel Krahmer
Citations in the corpus
- Ethical and social risks of harm from Language Models (2021, background): “Leveraging language processing tools and large public datasets to infer private traits is an active area of research (Kosinski et al., 2013; Park et al., 2015; Quercia et al., 2011; Youyou et al., 2015).”
- Survey of the State of the Art in Natural Language Generation: Core tasks, applications and evaluation (2017, background): “…phenomenon, the linguistic features of personality are only indirectly reflected in speaking or writing (a hypothesis underlying much work on detection of personality and other features in text, including Oberlander & Nowson, 2006; Argamon et al., 2007; Schwartz et al., 2013; Youyou et al., 2015).”
- Taxonomy of Risks posed by Language Models (2022, background): “Leveraging language processing tools and large public datasets to infer protected and other personal traits is an active area of research [107, 140, 146, 200], despite serious ethical concerns [6, 185].”
- Countering Misinformation and Fake News Through Inoculation and Prebunking (2021, background): “There is evidence that knowledge of 300 Facebook “likes” is sufficient to infer a user’s personality with greater accuracy than their spouse (Youyou et al., 2015).”
- Psychological targeting as an effective approach to digital mass persuasion (2017, background): “We chose these two because they show strong associations with Facebook Likes (14) and have been frequently investigated in the context of consumer preferences and persuasive communication (e.g., ref.”
- Psychology Meets Machine Learning: Interdisciplinary Perspectives on Algorithmic Job Candidate Screening (2018, background): “For instance, there is evidence that computer-based personality judgments based on digital cues are more accurate than those made by humans (Youyou et al. 2015).”
- Studying User Income through Language, Behaviour and Affect in Social Media (2015, background): “Crowdsourcing perceived attribute annotations—collecting the subjective impressions workers get when reading the content of user profiles—is not trivial [28, 29].”
- Citizens Versus the Internet: Confronting Digital Challenges With Cognitive Tools (2019, background): “, Facebook likes) can be more accurate than judgments made by relatives and friends (Youyou et al., 2015), and just 300 likes are sufficient for an algorithm to predict users’ personalities more accurately than their own spouses can (Youyou et al.”
- Collective aspects of privacy in the Twitter social network (2018, background): “Facebook has also been shown to be extremely informative of user personal information, including sexual orientation [17], romantic partnerships [18], and a wealth of other private attributes that can be inferred from user “likes” [19, 20].”
- Analyzing Biases in Human Perception of User Age and Gender from Text (2016, background): “With the Internet a substantial part of daily life, users leave enough footprints which allow algorithms to learn a range of individual traits, some with even higher accuracy than the users’ own family (Youyou et al., 2015).”