Emotion theories: James-Lange, Cannon-Bard, Schachter-Singer, appraisal theory

Intuition Beginner

What comes first — the feeling or the bodily reaction? William James argued in 1884 that we feel afraid because we run, not the other way around. "We feel sorry because we cry, angry because we strike, afraid because we tremble." This James-Lange theory says bodily changes cause emotional feelings: perceive the bear, the body reacts, the brain reads the body, and that reading is the fear.

Walter Cannon disagreed. He argued the brain sends signals to the body and the conscious feeling at the same moment — you do not need bodily feedback to feel fear. This Cannon-Bard theory treats arousal and experience as parallel outputs. The signal goes to the cortex (feeling) and to the body (arousal) simultaneously, neither causing the other.

Stanley Schachter and Jerome Singer proposed a compromise. The body supplies general arousal, but the mind labels it as a specific emotion using the surrounding situation. This two-factor theory says a racing heart could be fear, anger, or love depending on context. Arousal is generic; the cognitive label is what makes the emotion specific.

Appraisal theory, developed by Richard Lazarus, goes further: how we interpret a situation determines our emotional response. A bear in a zoo is not frightening; a bear in your tent is. The same bodily arousal becomes different emotions because the meaning we assign differs. This unit traces how these four answers evolved and what the modern evidence now says.

Visual Beginner

The four theories differ mainly in what they put first and whether they treat arousal and feeling as separate. The table below maps the core claim and the main objection for each.

Theory	Causal order	Core claim	Main objection
James-Lange	Stimulus → body → feeling	Bodily changes cause the emotional experience	Body reacts too slowly; arousal is too similar across emotions
Cannon-Bard	Stimulus → (body ∥ feeling)	Arousal and feeling are simultaneous, independent	Ignores how context and cognition shape the emotion
Schachter-Singer	Stimulus → arousal + label → feeling	General arousal plus a cognitive label equals emotion	The original study is hard to replicate cleanly
Lazarus	Stimulus → appraisal → feeling	Cognitive appraisal of meaning generates the emotion	Some responses are too fast for any appraisal

These theories are not mutually exclusive. Modern research treats emotion as a blend of bodily arousal, cognitive appraisal, and social context acting together. The debates that follow refine rather than replace this core picture.

Worked example Beginner

Imagine your heart is pounding before a big presentation. Your palms sweat, your breathing quickens. Is this fear, excitement, or anger? The four theories give four different answers about what is happening inside you.

James-Lange. The pounding heart is the fear. Your brain reads your racing body and labels the state accordingly. Change the body and you change the feeling.

Cannon-Bard. Your brain registered the threat and triggered the pounding heart and the feeling of fear at the same moment. One does not wait for the other.

Schachter-Singer. The pounding heart is just arousal. Your mind scans the room for an explanation and labels the arousal "fear" (or "excitement") from the available cues. The body alone is ambiguous.

Lazarus. You appraised the talk as a threat before the pounding began — perhaps because of past failures or social pressure. That appraisal is what generated the emotion and the arousal that came with it.

The same racing heart maps to different feelings across situations — a first date, a fight, a finish line. No single theory captures all of it. Each isolates one piece of a process that, in real life, runs together.

Check your understanding Beginner

Formal definition Intermediate

An emotion is a coordinated, multi-component response comprising three parts:

1. Subjective experience — the felt quality (what it is like to be afraid, angry, joyful).
2. Physiological response — autonomic and hormonal changes (heart rate, skin conductance, cortisol, muscle tension).
3. Expressive and behavioural response — facial configuration, posture, vocal tone, and an action tendency (approach, withdrawal, attack, freeze).

The four canonical theories differ in the causal ordering they impose on these components. Let $S$ denote the stimulus, $P$ the physiological response, $E$ the subjective emotional experience, $A$ an undifferentiated arousal state, $L$ a cognitive label, and $\mathrm{App}(\cdot )$ an appraisal function. The theories then correspond to four causal chains:

$$\text{James-Lange:}S\to P\to E$$ $$\text{Cannon-Bard:}S\to (P\Vert E)$$ $$\text{Schachter-Singer:}S\to (A+L)\to E$$ $$\text{Lazarus:}S\to \mathrm{App}(S)\to (P,E)$$

The symbols $\Vert$ and $+$ are notational conveniences, not formal operators. $\Vert$ marks parallel, independent generation; $+$ marks a joint requirement (both arousal and a label are needed). The point of the notation is to make the disagreement precise: the theories disagree about which arrow comes first and whether arousal and experience are coupled. Each theory is testable insofar as it predicts a different ordering that should be disruptable by intervening at a different stage.

A useful way to read the four chains as a progression: James-Lange puts the body first; Cannon-Bard severs the body from the feeling; Schachter-Singer reunites them but makes the body non-specific and the label decisive; Lazarus promotes a cognitive appraisal process to the generating role. Each step absorbs the previous theory's insight while repairing its main defect.

Key model: the four canonical theories Intermediate

James-Lange theory

James (1884) and Carl Lange (1885) independently proposed that emotion follows bodily change. The chain is $S\to P\to E$: a stimulus triggers a patterned physiological response, and the brain's perception of that response constitutes the felt emotion. Each emotion is, on this view, the feeling of a specific bodily configuration — fear is the feeling of trembling, running, and a racing heart; anger is the feeling of clenched fists and a flushed face.

The strongest modern descendant of this idea is the facial feedback hypothesis: that the face feeds back to the brain and shapes the emotional experience. Strack, Martin, and Stepper (1988) famously reported that holding a pen between the teeth (activating a smile-like expression) made cartoons seem funnier than holding it between the lips (activating a frown-like expression). David Havas and colleagues extended the logic using botulinum toxin: temporarily paralysing the corrugator muscle (the frown muscle between the brows) slowed people's reading of sad sentences, consistent with the idea that blocking the bodily expression blunts the underlying feeling.

Cannon's objections remain serious. Bodily responses are slow (autonomic changes unfold over seconds, while feelings appear within milliseconds), and the autonomic signatures of different emotions overlap heavily. A racing heart accompanies fear, anger, excitement, and sexual arousal alike. If the brain reads the body to tell emotions apart, the body does not give it enough information to do so.

Cannon-Bard theory

Cannon (1927) and Philip Bard argued that the thalamus (and related emotion circuits) sends signals to the cortex and to the autonomic nervous system at the same time. The chain is $S\to (P\Vert E)$: the physiological response and the conscious feeling are parallel, independent outputs. Neither causes the other; both are caused by the central processing of the stimulus.

Cannon's critique of James-Lange drew on several lines of evidence. Artificially induced arousal (an adrenaline injection) does not by itself produce a specific emotion. Severing the spinal cord in cats did not abolish their emotional behaviour, even though it cut off bodily feedback to the brain. And the physiological changes associated with different emotions are too undifferentiated to serve as the basis for distinguishing them.

A later and more telling piece of evidence came from George Hohmann (1966), who interviewed people with spinal cord injuries. Those with higher lesions — and therefore less bodily feedback — reported changes in emotionality (less intense anger and fear, more "mental" feeling), but they did not stop feeling emotions entirely. This is double-edged: it refutes the strong James-Lange claim that bodily feedback is necessary for emotion, while leaving room for a weaker role in which the body modulates intensity.

Schachter-Singer two-factor theory

Schachter and Singer (1962) proposed that emotion requires two factors: undifferentiated physiological arousal plus a cognitive label that attributes the arousal to a source. The chain is $S\to (A+L)\to E$. The arousal is generic and cannot distinguish emotions on its own; the label, supplied by the surrounding context, does the discriminating work.

Their experiment is one of the most famous in psychology. Participants received an injection of epinephrine (adrenaline) or a placebo. Some were correctly informed about the drug's effects; others were misinformed or told nothing. Each participant then waited with a confederate who acted either euphoric (playing with paper, shooting hoops) or angry (tearing up a questionnaire and storming out). Uninformed participants — who felt arousal but had no ready explanation for it — tended to "catch" the confederate's mood, reporting euphoria or anger consistent with the social context. Informed participants, who could attribute their arousal to the injection, did not catch the mood.

The study has not aged cleanly. Marshall and Zimbardo (1979) failed to replicate the anger condition convincingly, and later critiques questioned the demand characteristics and the statistical analyses. The strong version of two-factor theory — that arousal is a blank slate awaiting any label — is not well supported. But the weaker core insight survives: the same arousal can be experienced as different emotions depending on how the situation is construed, and people sometimes misattribute arousal to the wrong source (as in the "suspension bridge" studies of attraction, Dutton and Aron 1974).

Lazarus appraisal theory

Richard Lazarus (1966, 1991) placed cognitive appraisal at the centre of emotion. The chain is $S\to \mathrm{App}(S)\to (P,E)$: the stimulus is not enough; the person must evaluate its significance, and that evaluation generates both the feeling and the bodily response. Lazarus insisted that appraisal can be rapid and nonconscious — it need not involve deliberate reflection — but he maintained that some evaluation of personal meaning is a necessary condition for emotion.

Lazarus distinguished two appraisal stages. Primary appraisal evaluates whether the situation is relevant to well-being and, if so, whether it is threatening, harmful, or benign-positive. Secondary appraisal evaluates coping resources — can I handle this, and what options do I have? The resulting emotion depends on both. A bear in a zoo is appraised as irrelevant; a bear in your tent is appraised as both threatening and beyond your coping resources, and the emotion is terror. Reappraisal — revising the evaluation as the situation develops — can change the emotion mid-stream.

Craig Smith and Phoebe Ellsworth (1987) factor-analysed the appraisal dimensions that differentiate emotions. Their framework identifies six dimensions along which situations are evaluated:

• Pleasantness — is this enjoyable or aversive?
• Anticipated effort — will this require exertion?
• Certainty — how predictable is the situation?
• Attentional activity — how much does this demand attention?
• Self-other responsibility / control — who is accountable, and who has agency?
• Situational control — how much can circumstances determine the outcome?

Different patterns across these dimensions map onto different emotions. Anger combines unpleasantness with other-responsibility; guilt combines unpleasantness with self-responsibility; sadness combines unpleasantness with low control and low effort.

Klaus Scherer's component process model pushes appraisal further into a sequence of stimulus evaluation checks that unfold over time: novelty, intrinsic pleasantness, goal significance, coping potential, and norm compatibility. Each check appraises a different aspect of the stimulus, and the emotion is the dynamic result of the sequence, not a single static evaluation. Scherer's model is the most fine-grained of the appraisal theories and connects appraisal to measurable physiological changes as the checks succeed one another.

A famous exchange in the 1980s set Lazarus against Robert Zajonc, who argued that affective preferences can arise too quickly for any appraisal — we simply like or dislike before we know why. Lazarus countered that even rapid appraisals involve a (possibly unconscious) evaluation of meaning. The modern consensus sits between them: some affective reactions are near-reflexive and minimally cognitive, while richly differentiated emotions depend heavily on appraisal.

Exercises Intermediate

Competing perspectives: basic emotions, universality, and construction Master

The basic-emotions framework

Paul Ekman's research from the 1960s onward argued that a small set of emotions — happiness, sadness, fear, anger, surprise, and disgust — are expressed through the same facial configurations in all human cultures, including preliterate societies such as the Fore of Papua New Guinea. Ekman later treated contempt as a possible seventh basic emotion, though the evidence is weaker and more contested than for the original six.

A related distinction is the Duchenne smile: a genuine smile that engages both the zygomatic major (pulling up the mouth corners) and the orbicularis oculi (crinkling the skin around the eyes). Non-Duchenne or "social" smiles engage only the mouth. People across cultures can distinguish the two at above-chance rates, and Duchenne smiles correlate with genuinely felt positive emotion more reliably than social smiles do. The distinction matters because it suggests the face carries information about the sincerity of an emotion, not just its category.

Neurocultural theory and the universality claim

Ekman's neurocultural theory (1972) holds that facial expressions are produced by universal, innate "affect programmes" but are filtered through cultural display rules — learned norms about when and to whom emotions may be shown. On this view the underlying programmes are universal; what varies is whether, and how, they are allowed to reach the face. This is meant to reconcile the cross-cultural consistency Ekman reported with the obvious fact that cultures differ in emotional expression.

The methodological critique

James Russell (1994) reviewed the cross-cultural evidence and argued that the apparent universality was substantially inflated by method. Most studies used a forced-choice format: participants were shown a face and asked to select an emotion from a short list. With six options, chance alone produces roughly seventeen percent agreement, and the format removes the option to say "none of the above" or to give a non-basic label. When researchers used free-response methods — letting participants label the expression in their own words — agreement rates dropped sharply, and dropped further still in cultures far from the Western, literate societies where the emotion-word lists were standardised.

Crivelli, Carlos, and Russell extended this critique in the field. In studies of the Trobriand Islanders of Papua New Guinea (Crivelli et al. 2017), the match between Western-predicted facial expressions and local emotion attributions was much weaker than Ekman's framework predicts. Islanders often matched the same face to different emotions, or matched the same emotion to different faces, in ways inconsistent with a strict six-category universal scheme. The cross-cultural inference from a face to a discrete emotion is real but weaker, and more context-dependent, than the basic-emotions literature often claims.

The theory of constructed emotion

Lisa Feldman Barrett's theory of constructed emotion (2017) is the most radical alternative to the basic-emotions and discrete-programme view. On this account, emotions are not triggered responses produced by dedicated circuits. They are constructed predictions that the brain assembles on the fly from three ingredients:

• Interoception — the brain's ongoing model of the body's internal state (heart, lungs, gut, metabolism).
• Core affect — a continuous space of valence (pleasant to unpleasant) and arousal (low to high).
• Conceptual knowledge — learned categories, shaped by language and culture, that the brain uses to make meaning of bodily states.

The same state of unpleasant, high-arousal affect can be categorised as anxiety, anger, or determination depending on the concepts the brain brings to bear. There are no dedicated fear circuits or anger circuits waiting to fire; instead, population coding and degeneracy — many different neural configurations producing the same outcome — mean that the same emotion category is realised by different brain states on different occasions.

The empirical case rests heavily on the absence of the regularities a basic-emotions view predicts. Clark-Polner and colleagues reviewed neuroimaging work and found no consistent, emotion-specific brain-activation pattern that holds across studies, contexts, and individuals. Regions canonically associated with single emotions — the amygdala with fear, the insula with disgust — are multifunctional: the amygdala is involved in novelty detection, salience, and social judgement as much as in fear, and ultra-high-field fMRI shows that these regions contain intermixed, distributed populations rather than emotion-specific modules. If emotions were natural kinds with dedicated circuits, the circuits should be findable; the constructionist argument is that decades of looking have not found them.

What is at stake

The disagreement is not a taxonomic quibble. If the basic-emotions view is right, emotion categories carve nature at its joints: there is a real, biologically fixed set of emotional kinds, and research should aim to map their circuits and signatures. If the constructionist view is right, emotion categories are tools the brain and a culture use to make sense of bodily states, and research should aim to understand the construction process — interoception, categorisation, and the social shaping of emotional concepts.

Most working scientists occupy a middle position. Some cross-cultural consistency in expression and recognition is real, particularly for a few high-consensus categories. But the consistency is weaker, more context-dependent, and more shaped by culture than the original universality claim allowed. The raw materials — approach and withdrawal, pleasure and displeasure, calm and arousal — are biologically grounded. The rich taxonomy of named emotions is, to a substantial degree, constructed.

Connections: from appraisal to affective neuroscience Master

Affective neuroscience

Jaak Panksepp's affective neuroscience identifies seven primary emotional systems in the mammalian brain, each a coherent cross-species circuit studied through electrical stimulation and pharmacology: SEEKING (anticipatory eagerness and reward pursuit), RAGE (frustration and defensive aggression), FEAR (threat and escape), LUST (sexual drive), CARE (nurturance), PANIC/GRIEF (separation distress), and PLAY (rough-and-tumble social joy). Panksepp capitalises the labels to mark them as systems rather than folk-emotion words. His programme argues that these circuits are evolutionarily ancient and shared across mammals, and he tied them to psychiatric drug discovery — for instance, proposing that the PANIC/GRIEF system underlies the pain of depression and social loss, and that drugs acting on it might reach depression's emotional core in ways monoamine reuptake inhibitors do not.

Panksepp's framework sits in tension with constructionism: he treats the seven systems as real, hardwired affect programmes, whereas Barrett denies dedicated circuits for discrete emotions. One reconciliation is that Panksepp's systems operate at a more primitive, subcortical level than the emotion categories people name. The SEEKING and PLAY systems may be biologically basic, while the named emotions built atop them (curiosity, contentment, ennui) are culturally elaborated.

Interoception and emotion

A. D. Craig's (2002, 2009) work on interoception locates the neural map of the body's internal state in the posterior-to-mid insula, with a re-representation in the anterior insula that contributes to the conscious feeling of the body. The insula is therefore a prime anatomical candidate for the bodily-input stage of any James-Lange-style or constructionist account. People vary in interoceptive accuracy — the ability to detect their own heartbeat without taking a pulse — and this accuracy predicts the intensity of subjective emotional experience. Individuals who are better at sensing their bodies tend to report emotions as more intense, consistent with the idea that the felt emotion is partly a read-out of bodily state.

Affect labeling and emotion regulation

Matthew Lieberman and colleagues (2007) showed that affect labeling — putting a feeling into words ("this is anger") — reduces amygdala activity and increases activity in the right ventrolateral prefrontal cortex, a region associated with cognitive control. The act of naming an emotion appears to recruit prefrontal mechanisms that dampen the limbic response. This connects appraisal theory to regulation: the cognitive act of labelling is not merely descriptive but causally alters the emotional state.

James Gross's process model of emotion regulation organises regulatory strategies by where they act in the emotion-generation sequence. Situation selection and modification change the input; attentional deployment redirects focus (distraction, rumination, mindfulness); cognitive change reappraises the meaning (the strategy Lieberman's affect labeling draws on); and response modulation alters the output (expressive suppression, substance use, deep breathing).

The evidence favours reappraisal. Cognitive reappraisal reduces both the subjective experience and the physiological signature of negative emotion, and people who habitually reappraise show better well-being and lower depression rates. Expressive suppression — hiding the outward signs of an emotion — reduces the visible expression but leaves the subjective feeling largely intact and carries a physiological cost: elevated sympathetic activation and impaired memory for the event (Gross 1998; Butler et al. 2003). The cultural picture is nuanced: suppression is more normative and less socially costly in East Asian contexts that value emotional restraint (Butler, Lee, and Gross 2007), so the "maladaptive" verdict is moderated by the cultural meaning of the strategy.

Embodiment and facial feedback revisited

Paula Niedenthal's work on embodied emotion argues that understanding an emotion — in oneself or another — involves partially simulating the associated bodily state, with mirror-neuron systems and somatosensory cortex playing a role. On this account, recognising a smile involves activating (covertly) the neural machinery of smiling. Embodied cognition here provides a modern home for James-Lange intuitions: the body is not an output stage but part of the representational machinery of emotion itself.

The facial feedback hypothesis has been retested under modern standards. The Strack pen-in-mouth finding did not survive a large preregistered, multi-site replication (Wagenmakers et al. 2016). But the broader claim was salvaged by the Many Smiles Collaboration (Coles et al. 2022), a preregistered pooled analysis across many labs that found a small but statistically reliable effect of facial-mimicry manipulations on self-reported emotional experience — small enough to matter mainly at scale, but not zero. The botulinum-toxin literature offers convergent evidence from the opposite direction: Havas and colleagues (2010) found that paralysing the frown muscle slowed the reading of sad sentences, and Finzi and Schultz have reported that botulinum toxin injected into frown-related muscles can produce temporary relief from depression, a finding now tested in randomised trials. If blocking the body's expression blunts the feeling, the body is doing some of the work — just not as much, or as cleanly, as James claimed.

Stress, health, and the next unit

The appraisal framework is the bridge to health psychology. The same event produces very different physiological tolls depending on whether it is appraised as a threat (demands exceed resources) or a challenge (resources are adequate). Lazarus's primary and secondary appraisal are the direct ancestors of the stress-appraisal model that the next unit, 29.11.03 on stress and health, develops. Emotion regulation — reappraisal versus suppression — also predicts long-term cardiovascular and immune outcomes, which is why emotion theory and stress theory cannot be cleanly separated.

Historical and philosophical context Master

From James to Barrett

The modern study of emotion begins with William James's 1884 essay "What Is an Emotion?" James inherited a philosophical tradition, running from Descartes's Passions of the Soul (1649), that treated emotions ("passions") as disturbances of the soul produced by the body. James inverted this: rather than the mind being disturbed by the body, the mind's reading of the body just is the emotion. This was a radically embodied theory, and it set the terms of debate for the next century.

Cannon's 1927 critique was the next watershed. By arguing that the brain generates arousal and feeling in parallel, Cannon shifted emotion theory inward — from the body to the brain — and his framing dominated the mid-twentieth century, an era when behaviourism had little use for inner states at all. The cognitive revolution of the 1960s and 1970s restored inner states to respectability, and Schachter-Singer (1962) was among the first to give cognition a starring role: arousal is meaningless until the mind labels it.

Lazarus's appraisal theory, developed from the 1960s through his 1991 Emotion and Adaptation, made cognition not just a labeller of arousal but the generative engine of emotion. His debate with Zajonc over whether cognition is necessary for emotion was, in retrospect, a dispute about how much processing counts as "cognitive." Once rapid, nonconscious appraisal was granted the status of cognition, the extreme versions of both positions softened.

The constructionist tradition, now most fully developed by Barrett, descends from a different lineage: the dimensional theories of Wundt and Russell (emotion as variation along valence and arousal dimensions) and the psychological constructionism of the 1960s (Schachter's own two-factor theory is arguably an early constructionist view). Barrett's contribution is to ground construction in predictive-coding and interoceptive neuroscience, arguing that the brain is fundamentally a prediction machine that constructs categories — including emotion categories — to explain bodily states.

Natural kinds versus constructed categories

The philosophical fault line beneath this history is the question of whether emotion categories are natural kinds — real, biologically fixed divisions in nature, like chemical elements — or constructed categories — useful but historically contingent groupings, like biological species or, more provocatively, like colours.

If emotions are natural kinds, then there is a fact of the matter about how many basic emotions there are, what their circuits are, and where their boundaries lie. Research is a discovery problem. If emotions are constructed categories, then the boundaries reflect the purposes of the categoriser: a culture that needs to distinguish schadenfreude from joy will have words and concepts for both, while a culture that does not will not. Research is partly a discovery problem and partly an investigation of how brains and cultures build categories at all.

This is not a debate that evidence alone can settle, because the two sides disagree about what would count as evidence. A constructionist points to variability; a basic-emotions theorist points to the regularities that survive the variability. Both are looking at the same data and drawing different conclusions about its significance — a pattern familiar from many mature scientific disputes.

The colonial framing of "universal" emotion

There is a further, methodological wrinkle. The classic cross-cultural studies asked, in effect, "Do non-Western people recognise our emotion categories from facial expressions?" The Western categories were treated as the standard against which other cultures were measured. This framing inherits a colonial assumption: that Western taxonomies are the reference point and other cultures are the test case.

This does not invalidate the cross-cultural evidence, but it does skew it. Studies that begin by asking what emotion categories a given culture actually uses — rather than testing recognition of a fixed Western list — regularly uncover categories that do not map onto English. The Ifaluk fago, the German Schadenfreude, the Japanese amae are not exotic curiosities; they are data about the granularity of emotional life that the Western list flattens. Constructionism, by treating all emotion categories (including English ones) as constructed, re-frames the comparison on equal footing: no culture's categories are privileged as the standard, and the research question becomes how categories are built and shared, not which ones are "real."

Bibliography Master

1. James, W., "What Is an Emotion?," Mind 9(34) (1884), 188-205.

2. Lange, C. G., Om Sindsbevaegelser: En Psyko-Fysiologisk Studie (Kromar, 1885); trans. in K. Dunlap (ed.), The Emotions (Williams & Wilkins, 1922).

3. Cannon, W. B., "The James-Lange Theory of Emotions: A Critical Examination and an Alternative Theory," American Journal of Psychology 39(1/4) (1927), 106-124.

4. Bard, P., "A Diencephalic Mechanism for the Expression of Rage with Special Reference to the Sympathetic Nervous System," American Journal of Physiology 84 (1928), 490-515.

5. Schachter, S. and Singer, J. E., "Cognitive, Social, and Physiological Determinants of Emotional State," Psychological Review 69(5) (1962), 379-399.

6. Lazarus, R. S., Psychological Stress and the Coping Process (McGraw-Hill, 1966).

7. Hohmann, G. W., "Some Effects of Spinal Cord Lesions on Experienced Emotional Feelings," Psychophysiology 3(2) (1966), 143-156.

8. Ekman, P., "Universals and Cultural Differences in Facial Expressions of Emotion," in J. Cole (ed.), Nebraska Symposium on Motivation (University of Nebraska Press, 1972), 207-283.

9. Dutton, D. G. and Aron, A. P., "Some Evidence for Heightened Sexual Attraction Under Conditions of High Anxiety," Journal of Personality and Social Psychology 30(4) (1974), 510-517.

10. Lazarus, R. S., Emotion and Adaptation (Oxford University Press, 1991).

11. Ekman, P., "An Argument for Basic Emotions," Cognition and Emotion 6(3/4) (1992), 169-200.

12. Russell, J. A., "Is There Universal Recognition of Emotion From Facial Expression? A Review of the Cross-Cultural Studies," Psychological Bulletin 115(1) (1994), 102-141.

13. Strack, F., Martin, L. L., and Stepper, S., "Inhibiting and Facilitating Conditions of the Human Smile: A Nonobtrusive Test of the Facial Feedback Hypothesis," Journal of Personality and Social Psychology 54(5) (1988), 768-777.

14. Gross, J. J., "Antecedent- and Response-Focused Emotion Regulation: Divergent Consequences for Experience, Expression, and Physiology," Journal of Personality and Social Psychology 74(1) (1998), 224-237.

15. Marshall, G. D. and Zimbardo, P. G., "Affective Concomitants and Consequences of Maturation and Information in the Emotional Experience of Fear," (unpublished manuscript, Stanford University, 1979).

16. Lazarus, R. S., "Cognition and Motivation in Emotion," American Psychologist 46(4) (1991), 352-367.

17. Smith, C. A. and Ellsworth, P. C., "Patterns of Cognitive Appraisal in Emotion," Journal of Personality and Social Psychology 48(4) (1985), 813-838.

18. Scherer, K. R., "Appraisal Theory," in T. Dalgleish and M. J. Power (eds.), Handbook of Cognition and Emotion (Wiley, 1999), 637-663.

19. Zajonc, R. B., "Feeling and Thinking: Preferences Need No Inferences," American Psychologist 35(2) (1980), 151-175.

20. Craig, A. D., "How Do You Feel? Interoception: The Sense of the Physiological Condition of the Body," Nature Reviews Neuroscience 3(8) (2002), 655-666.

21. Panksepp, J., Affective Neuroscience: The Foundations of Human and Animal Emotions (Oxford University Press, 1998).

22. Lieberman, M. D., Eisenberger, N. I., Crockett, M. J., Tom, S. M., Pfeifer, J. H., and Way, B. M., "Putting Feelings Into Words: Affect Labeling Disrupts Amygdala Activity in Response to Affective Stimuli," Psychological Science 18(5) (2007), 421-428.

23. Butler, E. A., Egloff, B., Wilhelm, F. H., Smith, N. C., Erickson, E. A., and Gross, J. J., "The Social Consequences of Expressive Suppression," Emotion 3(1) (2003), 48-67.

24. Butler, E. A., Lee, T. L., and Gross, J. J., "Emotion Regulation and Culture: Are the Social Consequences of Emotion Suppression Culture-Specific?," Emotion 7(1) (2007), 30-48.

25. Havas, D. A., Glenberg, A. M., Gutowski, K. A., Lucarelli, M. J., and Davidson, R. J., "Cosmetic Use of Botulinum Toxin-A Affects Processing of Emotional Language," Psychological Science 21(7) (2010), 895-900.

26. Niedenthal, P. M., "Embodying Emotion," Science 316(5827) (2007), 1002-1005.

27. Wagenmakers, E.-J., Beek, T., Dijkhoff, L., Gronau, Q. F., et al., "Registered Replication Report: Strack, Martin, & Stepper (1988)," Perspectives on Psychological Science 11(6) (2016), 917-928.

28. Crivelli, C., Carlos, G., and Russell, J. A., "The Fear Is a Happy Man' Challenging the Universal Recognition of Fear From Faces in the Trobriand Islands," Proceedings of the Annual Meeting of the Cognitive Science Society (2017).

29. Barrett, L. F., How Emotions Are Made: The Secret Life of the Brain (Houghton Mifflin Harcourt, 2017).

30. Coles, N. A., March, D. S., Marmolejo-Ramos, F., et al., "A Multi-Lab Test of the Facial Feedback Hypothesis by the Many Smiles Collaboration," Nature Human Behaviour 6 (2022), 1731-1742.

31. Clark-Polner, E., Wager, T. D., Satpute, A. B., and Barrett, L. F., "Neural Vulnerability to the Affective and Social Consequences of Sleep Loss: Is There a Trait Marker?," in L. F. Barrett, M. Lewis, and J. M. Haviland-Jones (eds.), Handbook of Emotions (4th ed., Guilford, 2016) — and related reviews on emotion-specific brain patterns.

32. Finzi, E. and Schultz, W. M., "Rapid and Sustained Improvement in Depression from a Single Dose of Botulinum Toxin A: A Double-Blind Study," (Finzi-Schultz, ongoing randomised trial work on botulinum toxin for depression).

33. Myers, D. G. and DeWall, C. N., Psychology, 13th ed. (Worth, 2021), Ch. 10.

34. Gleitman, H., Gross, J., and Reisberg, D., Psychology, 8th ed. (Norton, 2011), Ch. 6.