Ekman’s Basic Emotions: An S-Tier Behavioral Designer’s Guide

In 1967 a young psychologist flew to the highlands of Papua New Guinea carrying a stack of photographs and a question that most anthropologists of the era considered already answered. The answer everyone agreed on was that emotional expressions are learned, like words. A smile in one culture might mean something entirely different in another. Paul Ekman went looking for a culture remote enough to test that assumption, and he found the Fore: a people who had been living in the Stone Age a decade earlier, who had seen almost no outsiders, no magazines, no films, no photographs of foreign faces.

He showed them pictures of American faces and told them simple stories. Your child has died. You are about to fight. You see something that smells rotten. Then he asked them to point to the face that fit the story. They pointed to the same faces an American would have pointed to. He reversed it, asked the Fore to make the face, photographed them, and brought those photos home to American college students who had never heard of the Fore. The students read the New Guinean faces correctly too.

That fieldwork became the most famous claim in the science of emotion: a handful of feelings are not learned but inherited, written into the muscles of the human face the same way in every culture on Earth. Anger, fear, disgust, sadness, surprise, happiness. Later, contempt. Ekman called them basic emotions, and for half a century they shaped everything from psychiatry to airport security to the way Pixar designs a character’s face.

I care about this theory for a reason that has nothing to do with anthropology. If a small set of emotions really does show up involuntarily on the face, then the face is the closest thing a designer will ever get to an honest readout of what an experience did to a person, captured a half-second before their conscious mind can dress it up for a survey. Let me walk you through the theory, the expedition that built it, the serious case against it, and the crosswalk that turns a list of seven faces into a design tool for choosing which emotion you actually want to manufacture.

Speed Run Notes

• Ekman’s basic emotions are six discrete, biologically built-in feelings with signature faces: happiness, sadness, anger, fear, surprise, disgust. He later added contempt as a contested seventh.
• The evidence is the 1971 New Guinea study: an isolated preliterate people matched and posed the same expressions as Westerners, which Ekman read as proof the expressions are inherited, not learned.
• Culture still rules the face through display rules: the expression is universal, but who may show it, to whom, and when is learned. Japanese viewers masked disgust with a smile when an authority watched.
• The theory is under heavy fire. Russell showed forced-choice methods inflated agreement; Barrett’s 2019 review concluded you cannot reliably read a specific emotion off a face.
• For design, the basic emotions are an output palette, not an input. The leaked face is the most honest early signal of which emotion a design produced, before the user rationalizes it.
• In Octalysis terms each face maps to the Core Drives that produce it. Surprise is the face of Core Drive 7; fear is the face of Core Drive 8. Disgust is your warning light that Black Hat went too far.

Author Credibility: Yu-kai Chou

Yu-kai Chou created the Octalysis Framework after studying gamification since 2003 — years before the term entered mainstream vocabulary. As a Human-Systems Architect & Behavioral Designer, his framework has been applied by LEGO, Microsoft, Porsche, Coca-Cola, Salesforce, and MrBeast, impacting over 1.5 Billion Users.

Chou has taught the Octalysis methodology at Harvard, Stanford, Yale, Tesla, Google, BCG, and IDEO.

His work has been cited by Harvard, Stanford, MIT, Forbes, Wall Street Journal, Wired, US Department of Energy, NIST, NSF, NCBI, US Department of Education, ClinicalTrials.gov, and Google Scholar — with 3,700+ more academic publications. Explore his books here.

What Are Ekman’s Basic Emotions?

Ekman’s basic emotions are a short list of feelings he argued are not products of culture but products of evolution: built into the species, expressed the same way on every human face, and triggered automatically by the right circumstances. The original list, the one that ended up in psychology textbooks worldwide, has six members. Happiness. Sadness. Anger. Fear. Surprise. Disgust.

The word “basic” carries a specific claim. Ekman did not mean these emotions are simple or common. He meant they are foundational in the way primary colors are foundational: a small set of irreducible states that the more complicated feelings are built out of. Nostalgia is not basic. It is a blend, sadness braided with a kind of warmth. Jealousy is not basic. It mixes fear, anger, and the threat of loss. The basic emotions are the raw pigments.

To earn a place on the list, Ekman argued in his 1992 paper “An Argument for Basic Emotions,” a candidate emotion had to clear several bars. It needed a distinctive, universally recognized facial signal. It needed to appear quickly and automatically, without deliberation. It needed an evolutionary story, some survival problem it once solved. And it needed to show up in other primates, suggesting it predates human culture entirely. Fear clears every bar. It has a signature face, fires faster than thought, kept our ancestors alive by readying them to flee, and you can read it on a startled monkey.

Most summaries skip the part that matters most to a designer: Ekman’s theory is fundamentally a claim about output. The basic emotion is an internal program, and the face is its readout. Pull the trigger and the same display fires whether you are in Manhattan or the New Guinea highlands, whether you want it to or not. That involuntary quality is what makes the face interesting to anyone who wants to know what is really happening inside a person, because the face often moves before the mind can intervene.

The Expedition That Made the Case

To understand why this theory mattered so much, you have to understand what it was arguing against. By the 1950s the dominant view in social science was that almost everything human was learned. Margaret Mead and the cultural relativists had spent decades documenting how differently cultures organize family, sex, status, and emotion. The reigning assumption was that a facial expression was a kind of local dialect: an American smile and a New Guinean smile might be as unrelated as the English and Fore words for “happy.”

Ekman started out agreeing with them. What changed his mind was film. His mentor, Silvan Tomkins, could watch silent footage of a person’s face and tell you what they were feeling with eerie accuracy, and he insisted the expressions were universal. Ekman set out to prove Tomkins wrong, and his first cross-cultural data did the opposite. In a 1969 study published in Science with Sorenson and Friesen, people across the United States, Brazil, Chile, Argentina, and Japan all matched the same emotion words to the same faces.

The obvious objection was that all those cultures watch the same movies. Hollywood could have taught the whole literate world a shared vocabulary of faces. That is what sent Ekman to the Fore of Papua New Guinea, a group that had never seen a film or a magazine. In the 1971 study with Friesen, “Constants Across Cultures in the Face and Emotion,” he told Fore adults and children short stories tuned to each emotion, then asked them to choose the matching face from photographs. He had others act out the stories with their own faces and photographed them.

The Fore matched the expressions Westerners made, and Western judges later matched the expressions the Fore made. There was one revealing wrinkle: the Fore reliably confused fear and surprise with each other, which would turn out to matter decades later. But the headline held. People with no shared media and no shared language were making and reading the same faces. For a generation of researchers, that settled it. Expressions are not a dialect. They are closer to a heartbeat.

Ekman did not stop at recognition. He wanted a way to describe a face that did not smuggle in an interpretation, so that a researcher in Tokyo and a researcher in California would code the same movement identically. With Friesen he spent years building the Facial Action Coding System (FACS), published in 1978 and built partly on the anatomical groundwork of the Swedish anatomist Carl-Herman Hjortsjö. FACS breaks any human expression into roughly 40 Action Units, each one a specific muscle movement. A genuine smile is not “happy”; it is Action Unit 6 plus Action Unit 12, the cheek raiser and the lip corner puller together. That objective vocabulary is now the hidden backbone of Pixar’s character rigs and of nearly every emotion-recognition algorithm shipping today.

The Six, Then Seven, One Face at a Time

Each basic emotion, in Ekman’s account, solves a problem and wears a signature. Walking through them one at a time is the fastest way to feel why the theory was so persuasive.

Happiness is the social glue and the reward signal. The honest version, the Duchenne smile, pulls the lip corners up and crinkles the muscles around the eyes at the same time. The eye crinkle is hard to fake on purpose, which is exactly why a genuine smile reads as genuine. It says approach me, we are safe together.

Sadness is the signal of loss and the request for support. The inner eyebrows pull up and together, the lip corners drop. It slows a person down and, crucially, it pulls care toward them from others. A sad face is among the hardest to fake convincingly, which is part of why it moves us when it is real.

Anger readies the body to remove an obstacle or a threat by force. The brows pull down and together, the lips press or square, the eyes fix and harden. Anger is a status and boundary signal as much as a combat signal. It announces that a line has been crossed.

Fear prepares escape. The eyes widen to take in more of the scene, the brows raise and draw together, the lips stretch back. Everything about the face is built to gather information fast and get the body ready to run. Fear is the emotion Ekman tied most tightly to survival, and the one with the clearest evolutionary logic.

Surprise is the briefest of all, often gone in well under a second. Brows shoot up, eyes go wide, jaw drops. Its job is to reset attention: something unexpected just happened, drop what you were doing and orient. Surprise frequently collapses into another emotion a moment later, into fear if the surprise is bad or delight if it is good, which is why it is so easy to confuse with fear.

Disgust began as a food-rejection reflex and grew into a moral one. The nose wrinkles, the upper lip pulls up, the whole face recoils as if from a smell. The same expression that fires at rotten meat fires at a betrayal or a cruelty, which tells you the brain recruited an old physical reflex to handle a new social problem.

Years later Ekman proposed a seventh, contempt, the only one with an asymmetric signature: one lip corner tightening and lifting on a single side. Contempt is the look of feeling superior to someone, of writing them off. It has weaker cross-cultural support than the original six, and not every researcher accepts it, which is worth remembering whenever a confident chart lists exactly seven universal emotions.

By the 1990s Ekman had widened the idea well past faces. In his 1999 work on basic emotions he proposed a larger family that includes amusement, pride in achievement, relief, contentment, embarrassment, guilt, shame, and sensory pleasure. These do not all have a single distinct facial signature, which is why the tidy “six faces” version is the one that survived in popular memory. By 2016, in the Atlas of Emotions he built with his daughter Eve Ekman under a commission from the Dalai Lama, he surveyed emotion scientists and reported that more than 70% agreed on five emotion families: enjoyment, anger, fear, sadness, and disgust. The number keeps moving, which is itself a clue about how settled the science really is.

The Thinkers Who Built It

The idea did not start with Ekman. It started with Darwin. In 1872 Charles Darwin published The Expression of the Emotions in Man and Animals, arguing that human expressions are inherited remnants of behaviors that once served a direct purpose, and that they are shared across cultures and even species. The book sold well and then was largely ignored by psychology for nearly a century, because the field that grew up after him was committed to learning as the engine of almost everything.

The bridge from Darwin to Ekman was Silvan Tomkins. In the early 1960s Tomkins published Affect Imagery Consciousness and argued that a small number of innate affects, each with its own facial display, are the primary motivational system of human beings. Tomkins listed nine affects, including interest, surprise, joy, anger, fear, distress, disgust, and shame, and later added contempt. He was a brilliant reader of faces and a poor experimentalist, and he knew it. He pointed two of his students, Paul Ekman and Carroll Izard, toward the cross-cultural test that would either confirm or kill the idea.

Ekman ran with it for the next forty years, and the trajectory of his later career is part of the story. He turned FACS toward deception, arguing that suppressed emotions leak out in microexpressions, involuntary flashes lasting a fraction of a second that betray a feeling a person is trying to hide. That work made him famous, inspired the television series Lie to Me, and led to consulting with the FBI and the agency that became the basis for the TSA’s behavior-detection program. As we will see, that final chapter is also where the theory’s most serious credibility problems live.

What Ekman Got Right

It is fashionable now to treat basic emotion theory as debunked, and that is lazy. Ekman got several things right that have held up, and a designer who throws out the whole framework loses the parts that are most useful.

He was right that some expressions are recognized far above chance across wildly different cultures. The effect is strongest for happiness and weakest for the negative emotions, but even the skeptics who reanalyze his data find recognition rates that random guessing cannot explain. Whatever the exact mechanism, a smile is not arbitrary the way the word “smile” is arbitrary.

He was right that expression and culture are both real and that display rules reconcile them. In a 1972 study, Japanese and American viewers watched a stomach-churning film. Alone, both groups made the same disgust faces. With a scientist in the room, the Japanese viewers masked their disgust with a polite smile while the Americans did not. The underlying expression was shared; the rule about whether to show it was learned. That two-layer model, a universal display governed by culturally specific rules, is one of the more durable ideas in the field.

He was right that FACS is enormously useful even if you reject everything else. The system measures faces without claiming to know what they mean, which is precisely why it survives the controversy intact. You can use FACS to study the face and stay completely agnostic about whether there are six basic emotions or none.

And he was right about leakage in a narrow sense. The body does betray what the mind tries to suppress, through the face and through the voice. Sauter, Eisner, Ekman, and Scott showed in a 2010 study that the Himba of remote Namibia and English speakers recognized each other’s nonverbal vocalizations, the scream of fear, the laugh of joy, the disgusted retch, well above chance. The signal is real. The argument is about how precise and how universal it is.

Where Basic Emotion Theory Falls Apart

The case against Ekman is not a fringe position. It is the current center of gravity in emotion science, and a designer who wants to use the theory honestly has to hold the criticisms in view. There are three that matter.

The Method Inflated the Answer

In 1994 the psychologist James Russell published a review in Psychological Bulletin arguing that the classic universality studies were rigged by their own method, not on purpose, but structurally. Participants were handed a posed, exaggerated photograph stripped of voice, body, and context, then asked to choose from a short list of emotion words. That forced-choice format does a lot of the work. If the only options are “fear, anger, disgust, sad, happy, surprise,” a face that is merely unpleasant gets pushed toward the nearest available label. When researchers let people answer in their own words, or offered “none of these,” agreement dropped, sometimes sharply. Russell’s point was not that recognition is zero; it was that the studies measured something narrower than the universal-emotion story claimed. Ekman replied in 1994 with new data he said answered the critique, and the two camps have been arguing ever since.

The Brain Has No Fingerprint for Anger

The deeper challenge comes from Lisa Feldman Barrett and her theory of constructed emotion. Her claim is that there is no built-in program for fear or anger waiting to fire. Instead the brain takes raw, undifferentiated body arousal and, in the moment, constructs an emotion out of it using concepts it learned from its culture and the situation it is in. On this view the face is not the readout of a fear program, because there is no single fear program to read. In 2019 Barrett and four other senior scientists published a sweeping review in Psychological Science in the Public Interest that examined the evidence behind reading emotion from faces. Their conclusion was blunt: a scowl is not a reliable expression of anger, a pout is not a reliable expression of sadness, and you cannot confidently infer a person’s emotional state from their facial movements alone. People scowl when concentrating and smile when uncomfortable. The face is too variable and too context-dependent to be the diagnostic instrument the popular theory promised.

Maybe There Are Four, Not Six

A third line of work questions the count itself. Rachael Jack and colleagues at Glasgow, in a 2014 Current Biology study, modeled how facial expressions unfold over time rather than as frozen snapshots. Early in the signal, fear and surprise looked nearly identical to viewers, and so did anger and disgust. The distinctions arrived later. Jack’s team argued this points to four biologically basic signals, not six: happiness, sadness, fear-surprise, and anger-disgust, with the finer splits layered on top to add social information. Notice that the two pairs that merge, fear with surprise and anger with disgust, are exactly the confusions Ekman’s own Fore participants made back in 1971. The crack was visible in the founding data the whole time.

What’s Really Happening Inside the Brain

The brain evidence is genuinely split, and that split is the honest answer rather than a footnote. For years the cleanest story for basic emotions came from lesion and imaging work that seemed to tie specific emotions to specific structures. Damage to the amygdala impaired the recognition and experience of fear; damage to the insula and nearby basal ganglia impaired disgust. Reviews like Calder, Lawrence, and Young’s in 2001 organized a great deal of evidence around the idea that different emotions ride on partly separable neural systems, which fit the basic-emotion picture well.

Then the meta-analyses arrived and complicated it. When researchers pooled hundreds of brain-imaging studies, they did not find one tidy region that lights up for anger and only anger, or fear and only fear. The amygdala turns out to be involved in far more than fear; it responds to salience and uncertainty of many kinds. Emotions appear to emerge from distributed networks that overlap heavily across different feelings, which is the pattern Barrett’s constructionist account predicts. The picture today is that there are real and partly distinguishable patterns of brain activity for different emotions, but nothing as clean as one circuit per feeling.

For a designer the practical lesson survives the debate intact. Whether a feeling is a built-in program or a construction assembled on the spot, by the time it reaches the face it has already passed through the fast, largely automatic machinery of the brain. The face moves before deliberation catches up. That is why the involuntary expression, and especially the microexpression that leaks before a person composes themselves, carries information that a considered answer on a survey does not. You do not need to win the universality argument to use that fact.

Basic Emotions vs Other Emotion Theories

Basic emotion theory is one answer to the question of where a feeling comes from, and it lands in a specific spot relative to its rivals.

Against the two-factor theory of emotion, Ekman sits on the opposite side of a sharp divide. Schachter and Singer argued that the body produces one generic arousal and the mind reads the situation to decide which emotion it becomes. Ekman argued the opposite: that fear and anger and disgust each come with their own distinct bodily and facial signature from the start. The two-factor view says the label makes the emotion; the basic-emotion view says the emotion arrives already labeled by biology. Barrett’s constructionism is, in a sense, the two-factor idea grown up and armed with modern neuroscience.

Against appraisal theories, the relationship is more like a division of labor than a fight. Appraisal theory explains how an event gets evaluated, the relevance and the coping and the agency judgments, to produce a specific emotion. Basic emotion theory is more concerned with what the output looks like once it arrives and how it shows on the body. You can read appraisal as the input side and basic emotions as the output side of the same machine, which is roughly how a lot of working researchers treat them.

Against the dimensional view, the contrast is about discreteness. James Russell’s own circumplex model, the most influential alternative, holds that emotions are not separate boxes at all but points on two continuous axes: how pleasant and how aroused. On the circumplex, anger and fear are neighbors because both are unpleasant and high-arousal; there is no hard wall between them, only a region of feeling space. Ekman insists on the boxes. Modern evidence sits uncomfortably in between: there do seem to be somewhat distinct emotion categories, but they bleed into one another at the edges far more than a clean six-box model allows.

Basic Emotions in the Real World

Set the academic fight aside for a moment, because the theory has already reshaped entire industries, for better and worse.

The Workplace

FACS quietly runs through a lot of modern work. It trained a generation of negotiation and sales coaches to watch for the leaked microexpression, the flash of contempt across a buyer’s face that contradicts the polite “let me think about it.” Used well, that sensitivity makes a manager notice the fear behind an employee’s confident nod before a project quietly fails. Used badly, it turns into pseudoscience: the manager who is certain a direct report is lying because of a single raised eyebrow, when the science says that single cue carries far less information than that confidence assumes.

Marketing and User Experience

Ad agencies and product teams now wire up FACS-based software to webcams and record a viewer’s face frame by frame while they watch a commercial or move through an onboarding flow. The promise is a continuous, unfiltered read on whether the experience produces delight or friction, second by second, without waiting for a survey the customer will half-remember and rationalize. This is the most direct application of Ekman’s work to the work I do, and also the one that demands the most humility about what a face can and cannot tell you.

Education and Learning

The same expression-reading tools have moved into classrooms and tutoring software, watching for the furrowed brow of confusion or the slackening face of boredom so the system can adapt. The upside is a tutor that notices a student is lost before the student says so. The downside is a surveillance instrument pointed at children, justified by a science whose own founders argued about how reliable it is.

Health and Therapy

In clinical settings the discrete-emotion view has real teeth. Tools built on FACS help researchers study depression, autism, and chronic pain, where reading emotional signals is part of diagnosis and care. People on the autism spectrum often report that explicit instruction in the basic expressions helps them navigate social situations that neurotypical people read automatically. Here the theory earns its keep precisely because it offers a finite, nameable, teachable set.

The Emotion-Recognition Industry

The largest and most fraught application is the affect-recognition industry: a multibillion-dollar market selling software that claims to read emotion from faces at scale, deployed in hiring, in cars, in border control, in advertising. Almost all of it rests on Ekman’s basic emotions and on FACS. And it is exactly this industry that Barrett’s 2019 review aimed at, warning that the science does not support the confident claims being sold. When a system tells an employer that a job candidate “displayed anxiety,” it is converting a contested theory into a verdict about a person’s career. The gap between what the face actually signals and what the software claims to read is where the ethical danger lives.

The Elephant in the Room

Here is the thing almost no chart of the seven faces will tell you. The single most expensive real-world application ever built on Ekman’s basic emotions is also the one with the weakest evidence behind it, and the field looked away while it happened.

After the September 11 attacks, the TSA built a program called SPOT, Screening of Passengers by Observation Techniques, that trained airport officers to spot deception by reading travelers’ faces and behavior, grounded in Ekman’s microexpression work and FACS. By the time anyone audited it, the program had cost roughly 900 million dollars. In 2013 the Government Accountability Office reviewed it and was unusually blunt: across four meta-analyses covering more than 400 studies and sixty years, the human ability to detect deception from behavioral cues was at or barely above chance. The science did not support what the program claimed to do. A finite, universal set of involuntary faces is a genuine scientific finding. The leap from “the face leaks emotion” to “we can catch a liar or a terrorist by watching faces” was never earned, and a near-billion-dollar program ran on that unearned leap for years.

That is the elephant, and it is not a story about the TSA. It is a warning about a pattern, because the design industry is repeating it right now. The affect-recognition vendors selling webcam software that scores a job candidate’s “confidence” or flags a student’s “disengagement” are making the exact same leap: from the real but coarse fact that a face leaks emotional valence, to the unearned claim that a camera can read a specific named emotion accurately enough to decide someone’s future. Same theory, same overreach, higher resolution cameras. Lisa Feldman Barrett’s 2019 review was written in part to stop precisely this.

The uncomfortable truth a designer has to sit with is that the basic emotions are honest at low resolution and dishonest at high resolution. The face will tell you reliably that something pleasant or unpleasant just fired, faster and truer than a survey ever could. It will not reliably tell you it was specifically contempt rather than concentration, or specifically fear rather than surprise. Designing well means living in that gap: trusting the face as the most honest early signal you have, and refusing to pretend it is a verdict. The moment you treat a leaked expression as proof of a specific verdict about a specific person, you have left the science behind and joined the SPOT program.

How to Apply Basic Emotions with the Octalysis Framework

Most product teams I work with design the mechanic and then hope a good feeling shows up: ship the leaderboard, ship the streak, ship the loot box, and pray the user ends up delighted rather than anxious. They design the input and leave the emotional output to chance. Ekman’s basic emotions hand you the missing half of the equation, because they give you a finite, universal, involuntary palette of output states. There are not a thousand emotions to aim at. There is a short list of faces every human being on Earth makes.

The Octalysis Framework breaks all human motivation into 8 Core Drives. If basic emotions are the output, the Core Drives are the engines that produce them. Map one to the other and you can stop hoping for a feeling and start specifying it: name the face you want to see, then reverse-engineer the drives that create it.

The Crosswalk: Which Drive Produces Which Face

Surprise is the cleanest one-to-one map in the whole framework. Surprise is the momentary face of Core Drive 7 (CD7): Unpredictability & Curiosity. When a mystery box opens, when a feed serves something unexpected, when a game throws a twist, the brows jump and the eyes widen. If you want surprise on a user’s face, you reach for CD7. Just remember surprise is the most unstable emotion; it resolves into delight or fear within a second depending on what follows, which means CD7 by itself is a coin flip until you control the next beat.

Happiness is the White Hat enjoyment cluster firing together. Core Drive 2 (CD2): Development & Accomplishment produces the smile of a hard-won victory. Core Drive 3 (CD3): Empowerment of Creativity & Feedback produces the smile of making something work your own way. Core Drive 5 (CD5): Social Influence & Relatedness produces the smile of belonging. Genuine, eye-crinkling happiness in a product is almost never one drive; it is these three reinforcing each other.

Fear is the face of Core Drive 8 (CD8): Loss & Avoidance, sharpened by Core Drive 6 (CD6): Scarcity & Impatience. The countdown timer, the “your streak is about to break,” the “only 2 left” all aim at the same target: the widened eyes of someone afraid to lose. Fear converts in the short term, which is why Black Hat design leans on it. But a face stuck in fear is a face that eventually wants to escape, and escape from a product means churn.

Anger is rarely something you design on purpose, which makes it your most valuable diagnostic. Anger is the face of thwarted drives: blocked CD2 progress, violated Core Drive 4 (CD4): Ownership & Possession when something a user felt was theirs gets taken, or broken fairness inside CD5 when the social contract feels rigged. When you see anger, a Core Drive you activated has been frustrated rather than satisfied. The pay-to-win mechanic that enrages a free player is CD4 and CD5 curdling at once.

Sadness is loss after the fact, CD8 in its spent form. It is the face of the lost streak, the demoted rank, the abandoned save file. Sadness pulls for support, which is why the smartest win-back flows answer a sad churn with a hand extended rather than another scarcity threat.

Disgust and contempt are your alarm bells, and they are the most important faces in this entire crosswalk. Disgust is the recoil that says “this is manipulative.” Contempt is the one-sided sneer that says “this brand is beneath me.” When a Black Hat mechanic tips from motivating into exploiting, the first honest signal is not a complaint or a cancelled subscription. It is a flicker of disgust on the user’s face, often before they could tell you why. If your design is leaking disgust, your CD6 and CD8 engines have crossed the line from urgency into coercion.

The Design Instruction

Name the target face before you build the mechanic. Most teams cannot answer the question “what should be on the user’s face at this exact step,” and that inability is the root of a thousand flat, forgettable experiences. The basic emotions make the question answerable because the palette is finite. Decide that the reward screen should produce the eye-crinkling happiness of CD2 plus CD5, that the discovery moment should produce the wide-eyed surprise of CD7, and that no screen anywhere should leak disgust. Then build the drives that produce those faces, and treat any leaked disgust or contempt as a stop-the-line defect rather than a rounding error. You are no longer designing inputs and hoping. You are designing for the face you want to see.

Practical Steps for Designing the Face You Want to See

1. Storyboard the emotional output, not just the flow. For each key screen, write down the single basic emotion you want on the user’s face. If you cannot name it, the screen has no emotional intent, and a screen with no emotional intent is a screen users forget.
2. Map each target face back to its Core Drives. Want surprise? You need Core Drive 7 (CD7): Unpredictability & Curiosity. Want the happiness of accomplishment? You need Core Drive 2 (CD2): Development & Accomplishment and the feedback of Core Drive 3 (CD3): Empowerment of Creativity & Feedback. Build the engine that produces the face, not a generic “engagement” feature.
3. Watch the real face, not the survey. If you can run moderated tests, watch users’ faces during the moments that matter and note the involuntary expression before they speak. The half-second flash is more honest than the polite answer that follows it. The leaked face is your truest usability metric.
4. Treat disgust and contempt as stop-the-line signals. The moment a tester’s face recoils or sneers at a monetization screen, you have your earliest and most reliable warning that a Black Hat drive has tipped into exploitation. Do not explain it away. Redesign the screen.
5. Respect display rules in your own data. People mask. A user smiling politely through an onboarding flow in front of a moderator may be hiding frustration, exactly as Ekman’s Japanese viewers masked disgust before authority. Reduce the social pressure in your testing, or weight the unguarded micro-signals over the composed ones.
6. Stay humble about resolution, especially with software. If you use emotion-recognition tools, trust them to tell you roughly how pleasant and how aroused a user is, and distrust any claim that they have detected a specific named emotion with confidence. The science supports the coarse read, not the precise verdict.
7. Design the next beat, because surprise is unstable. Surprise becomes delight or fear within a second based on what happens next. Never trigger CD7 without controlling the moment that follows it, or you are gambling on which face you get.

Basic Emotions Were the Beginning, Not the End

Paul Ekman flew to New Guinea to settle whether feelings are universal, and the honest verdict half a century later is “partly, and less cleanly than he hoped.” The six faces are not the inviolable fingerprints the popular version promised. They are something more interesting: a reliable signal that an emotion of some valence and arousal has fired, captured before a person can compose themselves, paired with a much less reliable guess about exactly which emotion it is and why.

That is still the most honest early read on another person’s inner state that a designer will ever get, which is why the theory survives in practice even as it loses ground in theory. The face moves before the mind can edit it. For anyone building experiences, that involuntary half-second is a gift and a responsibility at once. Used to serve, it tells you the truth about whether your work delighted or quietly repelled the person in front of you. Used as a verdict, it becomes the SPOT program in miniature, a contested theory dressed up as certainty about a particular person. The instrument is the same; the only thing that changes is whether the person whose face you are reading would agree to be read.

If you take one thing from Ekman into your next design review, make it this: stop asking only what the user will do, and start asking what their face will do. Name the emotion you want, build the Core Drives that produce it, and watch the unguarded face to find out whether you succeeded. The mechanics are what you ship. The face is what you actually built.

Frequently Asked Questions About Ekman’s Basic Emotions

What are Ekman’s basic emotions?

Paul Ekman’s basic emotions are a small set of discrete emotions he argued are biologically built-in and recognized everywhere through distinct facial expressions. His core list is six: happiness, sadness, anger, fear, surprise, and disgust. He later added contempt as a seventh and, in the 1990s, proposed a longer family of emotions such as amusement, pride, relief, shame, and guilt that lack a single signature face.

What was Ekman’s New Guinea study?

In 1971 Paul Ekman and Wallace Friesen worked with the Fore people of Papua New Guinea, a preliterate group with almost no exposure to Western media. They told participants short emotion stories and asked them to pick the matching face from photographs of Westerners, and asked others to pose the emotion in a story. The Fore matched and produced the same expressions Westerners did, which Ekman argued was evidence that the expressions are universal rather than learned from culture.

How many basic emotions are there, six or seven?

Ekman’s original and most cited list is six: happiness, sadness, anger, fear, surprise, and disgust. He later added contempt as a seventh, though contempt has weaker cross-cultural evidence than the original six. Other researchers argue for four, merging fear with surprise and anger with disgust early in the signal, or for no fixed number at all.

What is the Facial Action Coding System (FACS)?

FACS is a system Ekman and Friesen published in 1978 that breaks any human facial movement into roughly 40 component Action Units, each tied to a specific muscle. It lets researchers describe a face objectively, without first deciding what emotion it shows. FACS is now the backbone of facial animation in film and of most emotion-recognition software.

Are facial expressions of emotion really universal?

Partly, and it is contested. Recognition of a few expressions, especially happiness, runs well above chance across very different cultures. But James Russell showed the classic studies leaned on a forced-choice method that inflated agreement, and Lisa Feldman Barrett’s 2019 review concluded that you cannot reliably read a specific emotion off a face. The honest summary is that faces signal emotion better than chance and far less precisely than the popular version of Ekman’s theory claims.

What are display rules?

Display rules are culturally learned rules about who may show which emotion to whom and when. Ekman’s neuro-cultural theory held that the underlying expression is universal but culture decides whether it gets suppressed, exaggerated, or masked. In his Japanese-American study, participants showed the same disgust faces alone, but the Japanese masked disgust with a smile when an authority figure was present.

What is the difference between Ekman’s theory and the theory of constructed emotion?

Ekman’s basic emotion theory says emotions are inherited programs with signature faces that fire across cultures. Lisa Feldman Barrett’s theory of constructed emotion says the brain builds each emotion in the moment from raw body arousal plus learned concepts and context, so there is no fixed fingerprint for anger or fear. The first treats the face as a readout of a built-in state; the second treats it as one variable that culture and context assemble into an emotion.

How do basic emotions apply to product and experience design?

Most teams design the mechanic and hope a good feeling follows. Basic emotion theory gives designers a finite, universal palette of output states to aim at, and the face is the most honest early signal of which one a design actually produced, often before a user can rationalize it on a survey. In Octalysis terms, each basic emotion maps to the Core Drives that tend to produce it, so a designer can name the target face first and reverse-engineer the drives that create it, while watching for the leaked disgust or contempt that warns a Black Hat design has tipped into manipulation.

References

• Darwin, C. (1872). The Expression of the Emotions in Man and Animals. London: John Murray.
• Tomkins, S. S. (1962–1963). Affect Imagery Consciousness (Vols. 1–2). New York: Springer.
• Ekman, P., Sorenson, E. R., & Friesen, W. V. (1969). Pan-cultural elements in facial displays of emotion. Science, 164(3875), 86–88.
• 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. (1972). Universals and cultural differences in facial expressions of emotion. Nebraska Symposium on Motivation, 19, 207–283.
• Ekman, P., & Friesen, W. V. (1978). Facial Action Coding System: A Technique for the Measurement of Facial Movement. Palo Alto: Consulting Psychologists Press.
• Ekman, P. (1992). An argument for basic emotions. Cognition and Emotion, 6(3–4), 169–200.
• Ekman, P. (1993). Facial expression and emotion. American Psychologist, 48(4), 384–392.
• Russell, J. A. (1994). Is there universal recognition of emotion from facial expression? A review of the cross-cultural studies. Psychological Bulletin, 115(1), 102–141.
• Ekman, P. (1994). Strong evidence for universals in facial expressions: A reply to Russell’s mistaken critique. Psychological Bulletin, 115(2), 268–287.
• Ekman, P. (1999). Basic emotions. In T. Dalgleish & M. Power (Eds.), Handbook of Cognition and Emotion (pp. 45–60). Chichester: Wiley.
• Calder, A. J., Lawrence, A. D., & Young, A. W. (2001). Neuropsychology of fear and loathing. Nature Reviews Neuroscience, 2(5), 352–363.
• Sauter, D. A., Eisner, F., Ekman, P., & Scott, S. K. (2010). Cross-cultural recognition of basic emotions through nonverbal emotional vocalizations. Proceedings of the National Academy of Sciences, 107(6), 2408–2412.
• Jack, R. E., Garrod, O. G. B., & Schyns, P. G. (2014). Dynamic facial expressions of emotion transmit an evolving hierarchy of signals over time. Current Biology, 24(2), 187–192.
• Barrett, L. F. (2017). How Emotions Are Made: The Secret Life of the Brain. New York: Houghton Mifflin Harcourt.
• Barrett, L. F., Adolphs, R., Marsella, S., Martinez, A. M., & Pollak, S. D. (2019). Emotional expressions reconsidered: Challenges to inferring emotion from human facial movements. Psychological Science in the Public Interest, 20(1), 1–68.
• U.S. Government Accountability Office. (2013). Aviation Security: TSA Should Limit Future Funding for Behavior Detection Activities (GAO-14-159). Washington, DC.

Related Reading

• The Two-Factor Theory of Emotion — the rival view that arousal needs a cognitive label before it becomes a feeling.
• Appraisal Theories of Emotion — how the mind evaluates an event to generate a specific emotion in the first place.
• The Octalysis Framework — the 8 Core Drives that serve as the engines producing each basic emotion.
• The Behavioral Framework Library — the full set of psychological models every behavioral designer should know.

WOULD YOU LIKE YU-KAI CHOU TO WORK WITH YOUR ORGANIZATION?