Paul Ekman

Introduction

Paul Ekman (15 February 1934 – 17 November 2025) established that certain emotional expressions are recognised across radically different cultures, gave psychology the instrument that made facial movement measurable, and then spent much of his later career trying to put limits on what his own findings could be used for, with limited success.

The finding is real and has been replicated: people in cultures as different as highland Papua New Guinea, urban Japan, and the United States label the same posed facial configurations with the same emotion words at rates well above chance. What that establishes is narrower than the conclusion usually drawn from it, and the gap between the two has cost real money and affected real people.

Ekman's second contribution is the Facial Action Coding System, an anatomically based taxonomy of facial movement that made expression objectively measurable for the first time. It remains the standard instrument in the field, and it is now embedded in commercial software that Ekman himself criticised.

He died at his home in San Francisco on 17 November 2025, aged 91, survived by his children Eve and Tom (New York Times, 2025). His daughter Eve Ekman, a contemplative social scientist, had taken over day-to-day leadership of the Paul Ekman Group as its executive director in 2023, two years before his death.

Early Life and Education

Ekman was born in Washington, D.C., and grew up in a Jewish family in New Jersey, apart from a brief period in California during the Second World War. His father was a paediatrician and his mother an attorney. His mother's suicide when he was fourteen is often cited as the origin of his wish to become a psychotherapist, a wish he later abandoned for research on the grounds that research could change institutions rather than one person at a time.

He enrolled at the University of Chicago in 1949, at fifteen, without having graduated from high school, in a period when the college admitted early. He completed three years there — his classmates included Susan Sontag, Mike Nichols, and Elaine May — and then transferred to New York University, where he took his B.A. in 1954. His first research project, supervised by Margaret Tresselt, tried to build a test of how people would respond to group therapy.

He was accepted into the clinical psychology programme at Adelphi University, received a predoctoral research fellowship from the National Institute of Mental Health in 1955, and wrote a master's thesis on facial expression and body movement. He completed a one-year internship at the Langley Porter Neuropsychiatric Institute and took his Ph.D. in clinical psychology at Adelphi in 1958, with a dissertation on how much information about emotion could be read from the body as opposed to the face.

The military draft intervened immediately. He served two years as first lieutenant and chief psychologist at Fort Dix, New Jersey, researching army stockades and psychological change during infantry basic training, and it was there that he concluded that research could make army routines more humane.

The question that shaped his career was posed to him in the early 1960s by his friend Silvan Tomkins: was expressive behaviour learned or innate? The standard position in American psychology at the time, strongly influenced by cultural anthropology, was that it was almost entirely learned, that gesture and expression were as arbitrary as language. Ekman set out to test it.

Academic and Professional Career

From 1960 to 1963 Ekman held an NIMH postdoctoral fellowship and worked as a research associate with Leonard Krasner at the Palo Alto Veterans Administration Hospital on the operant conditioning of verbal behaviour in psychiatric patients. In 1960 he met the anthropologist Gregory Bateson, who was on the staff at Palo Alto and who five years later handed him motion picture film shot in Bali in the mid-1930s to support cross-cultural work on expression and gesture.

The decisive study came in 1967 and 1968 among the Fore, an Indigenous group in the highlands of Papua New Guinea who had had limited contact with outsiders. Working with Wallace Friesen, Ekman asked Fore participants to select, from sets of photographed faces, the ones corresponding to brief scenarios — "he is looking at something that smells bad," "his child has died." The responses matched those of participants in the United States, Brazil, and Japan at rates far above chance, and in a complementary study American participants identified photographs of Fore people making expressions for the same scenarios at above-chance rates.

Ekman joined the University of California, San Francisco in 1972 and remained there for the rest of his career, retiring in 2004 as professor emeritus in the department of psychiatry. He received the American Psychological Association's Distinguished Scientific Contribution Award in 1991, and in 2002 the Review of General Psychology ranked him fifty-ninth among the hundred most eminent psychologists of the twentieth century.

The last quarter-century of his career was devoted to application rather than discovery. He collaborated over many years with the Dalai Lama on the classification of emotional states and on contemplative practice, co-authoring a book and webisode series on global compassion. With his daughter Eve he developed the online Atlas of Emotions, an interactive tool for building emotional vocabulary, and the Cultivating Emotional Balance training programme. He served as scientific adviser to Pixar's Inside Out. He had been a consultant to law enforcement and security agencies since the 1960s.

Major Contributions to Psychology

Basic emotions and universality

Ekman proposed that a small set of emotions — anger, disgust, fear, happiness, sadness, and surprise, with contempt and others added later — are biologically basic, with distinct, involuntary facial signatures that appear across cultures. The claim has two separable parts: that the expressions are produced universally, and that they are recognised universally. The evidence for recognition is considerably stronger than the evidence for production.

The Facial Action Coding System

Published in 1978 with Friesen, FACS is the contribution least likely to be superseded. Rather than coding faces by the emotion they supposedly express, it codes them by the underlying muscle movements — action units — with no reference to emotion at all. A configuration is described as the contraction of specific muscles, and only then mapped onto interpretation if desired.

This separation is what makes it scientifically usable. It allows researchers to measure what the face actually did without prejudging what it meant, which is precisely the distinction that the subsequent controversy turned on.

The Duchenne marker

Working from nineteenth-century anatomical work, Ekman distinguished the spontaneous smile from the deliberate one by whether the orbicularis oculi muscle contracts and produces crow's feet around the eyes, and catalogued eighteen distinguishable smiles. The distinction has become standard, though later work has questioned how reliably observers can code it in real time.

Microexpressions

Ekman coined the term for brief, involuntary facial movements lasting a fraction of a second, which he proposed leak concealed emotion. The idea originated in an observation he made while reviewing film of a psychiatric patient who had appeared calm in an interview but whose face briefly showed profound distress on slow-motion playback. The inference from microexpressions to deception detection is the step that has not been supported.

Major Publications

• Emotion in the Human Face (1972, edited with Wallace Friesen and Phoebe Ellsworth)
• Darwin and Facial Expression (1973) — a reconsideration of Darwin's Expression
• Unmasking the Face (1975, with Wallace Friesen)
• Facial Action Coding System (1978, with Wallace Friesen)
• The Face of Man (1980)
• Telling Lies: Clues to Deceit in the Marketplace, Politics, and Marriage (1985; 4th ed. 2009)
• Why Kids Lie (1992)
• Emotions Revealed (2003, 2nd ed. 2007)

Impact and Legacy

Ekman's influence outside psychology is unusually large. His work underlies the commercial emotion-recognition industry, software marketed for hiring, advertising, driver monitoring, and security, valued in the tens of billions of dollars. It underlies police and intelligence interview training. It inspired the television series Lie to Me, which was developed after the producer Brian Grazer read Malcolm Gladwell's 2002 profile of Ekman in The New Yorker.

Criticism — and it is substantial. Four separate objections have accumulated.

The method inflates agreement. Ekman's cross-cultural studies typically used forced-choice designs: participants saw a posed face and selected from a short list of emotion labels. Supplying the label set constrains the response space and raises agreement above what free labelling produces. Critics argue that this manufactures a substantial part of the universality it is meant to test. A related objection concerns the field condition itself: the Fore were not uncontacted, and by the late 1960s they had had decades of contact with missionaries, labour recruiters, and Australian administration.

The evidence for production is weak. Lisa Feldman Barrett and colleagues reviewed more than a thousand studies in 2019 and concluded that the mapping from emotion category to facial configuration is far looser than the basic-emotion account requires. Among urban populations, people scowl when angry about 30% of the time — above chance, but meaning that 70% of the time they are doing something else with their faces while angry. The scowl also appears during concentration, confusion, and dyspepsia. A 2012 study in PNAS found substantial cultural variation in the facial movements actually produced, and Rachael Jack's 2018 work found that Western and East Asian participants had similar concepts of pain faces but different ones for pleasure.

Deception detection does not work. This is the consequential failure. Ekman's microexpression work was the basis for SPOT — Screening Passengers by Observation Techniques — a programme the US Transportation Security Administration introduced in 2007 that trained officers to identify deception from faces and behaviour. It cost roughly $900 million. In 2013 the Government Accountability Office reviewed it and reported that across four meta-analyses covering more than 400 studies and sixty years of research, the human ability to detect deception from behavioural cues is at or barely above chance. The programme was criticised by scientists, by members of Congress, and by civil liberties organisations on both accuracy and racial-bias grounds.

The leap from face to verdict was never earned. The finding that the face carries some information about emotional valence is real. The claim that a face can be read with enough precision to decide whether a person is lying, is a security risk, or should be hired is not supported by any evidence, and Ekman said so himself, repeatedly and in print. But the distinction has not survived contact with the market, and the field largely looked away while a near-billion-dollar programme ran on it.

In fairness to the underlying science: some universality is established. Infants produce recognisable expressions before they could have learned them, congenitally blind individuals produce them, and cross-species comparisons support aspects of Darwin's original account. Elfenbein and Ambady's 2002 meta-analysis of more than a hundred studies confirmed above-chance cross-cultural recognition. The reasonable position is that universality is partial rather than absolute, that recognition is stronger than production, and that interpretation is heavily context-dependent, which is not what the popular version of Ekman's work says, and not what his commercial applications assume.

Learn More

• Charles Darwin — whose Expression of the Emotions (1872) posed the question Ekman tested
• Robert Plutchik — whose alternative model of emotion is also used commercially
• Body Language — the wider field, and its similar evidence problems
• Human Intelligence — for comparison, a construct where the measurement problems are better understood
• Carroll Izard — his collaborator on the MAX facial coding system, and the third architect of basic-emotion theory
• Lauri Nummenmaa — whose bodily maps of emotion located feelings in the body rather than only on the face

References

• Ekman, P., & Friesen, W. V. (1971). Constants across cultures in the face and emotion. Journal of Personality and Social Psychology, 17(2), 124-129.
• Ekman, P. (1992). An argument for basic emotions. Cognition and Emotion, 6(3-4), 169-200.