Laughter

Abstract

Laughter is a stereotyped, largely involuntary vocalization that humans produce in play, amusement, and social interaction. Although it is popularly tied to jokes, its scientific study treats it first as a social signal: Robert Provine's naturalistic observation showed that laughter occurs mostly in company rather than in solitude, punctuates the speaker's own speech at grammatical boundaries, and is only loosely coupled to humor. Cognitive psychology distinguishes spontaneous, emotionally driven laughter from volitional laughter produced on purpose, and the two differ in acoustics, in the brain systems that generate them, and in how listeners perceive them. Comparative work traces laughter to the play vocalizations of other mammals, and physiological work links shared laughter to endorphin and opioid release that supports social bonding. Three demonstrations present laughter as an acoustic bout, a social multiplier, and an authenticity signal listeners judge above chance.

Keywords: laughter, social vocalization, spontaneous and volitional laughter, laughter perception, social bonding

When people laugh, they rarely decide to. A laugh erupts, runs its course, and fades with little of the control that governs ordinary speech, which is the first clue that laughter is less a comment on the world than a signal exchanged within it. It is also far more social than its link to comedy suggests: most everyday laughter follows unremarkable conversational remarks rather than jokes, and it is produced overwhelmingly in the presence of others. Cognitive psychology approaches laughter with three questions. What is a laugh, as a physical and acoustic event, and how does the spontaneous laugh of genuine amusement differ from the laugh a person can produce at will? What does laughter do between people, as a signal that coordinates conversation, spreads through a group, and binds its members together? And where does laughter come from, in the brain that generates it and in the evolutionary history it shares with other animals? The sections that follow take these in turn. The recurring finding is that laughter is a social instrument, built on an ancient vocal play signal and only partly under voluntary control.

What Laughter Is

A laugh is a vocalization built from a rapid series of short, vowel-like notes — the familiar ha-ha-ha — produced on a single exhalation and repeated in bouts, each bout typically lasting a few seconds and riding on a distinctive respiratory and facial pattern. Willibald Ruch and Paul Ekman described this expressive pattern in detail, from the forceful, staccato exhalations and the contraction of the muscles around the eyes and mouth to the posture and the sometimes involuntary vocal character of the sound (Ruch & Ekman, 2001). Laughter is not a uniquely comic response: Robert Provine's program of naturalistic observation, which recorded where and when ordinary laughter actually occurs, found that most laughter in daily life follows banal conversational remarks rather than jokes or witticisms, and that it is woven into the structure of talk rather than reserved for moments of hilarity (Provine, 2000).

The single most important distinction in the modern study of laughter is between spontaneous and volitional laughter. Spontaneous laughter is the emotionally driven, largely involuntary laugh of genuine amusement or surprise; volitional laughter is produced deliberately, as a social nicety, a signal of affiliation, or a polite acknowledgment. The two are generated by partly separate systems and sound different: spontaneous laughter tends to be higher and more variable in pitch, longer, and less speech-like, while volitional laughter is more controlled and carries acoustic traces of the vocal apparatus used for speech. Nadine Lavan, Sophie Scott, and Carolyn McGettigan showed that authenticity modulates the acoustic, physiological, and perceptual properties of laughter, and that listeners pick up on these cues when judging whether a laugh is real (Lavan et al., 2016). This division maps onto an older observation, drawn from the study of smiling, between the genuine Duchenne display and its deliberate counterpart, and it organizes much of what follows (Gervais & Wilson, 2005). The first demonstration presents the acoustic structure of a laughter bout and lets the reader compare a spontaneous and a volitional profile.

A bout of laughter is a train of short notes on one exhalation. Choose the kind of laughter and the number of notes, and watch how the envelope changes.

Note. Spontaneous laughter tends to start loud and high and decay as the single exhalation runs out, with note-to-note variability; volitional laughter is flatter and more regular. The envelope here is schematic and generated from fixed formulae, not a recording (Ruch & Ekman, 2001; Lavan et al., 2016).

Laughter as a Social Signal

If laughter is only loosely about jokes, what is it for? Provine's answer was that laughter is fundamentally social. Recording thousands of instances of laughter in public, his group established the punctuation effect: laughter rarely interrupts the flow of speech but falls at its natural breaks, at the ends of phrases and sentences, as if a single system governs both the talk and the laughing that times itself around it (Provine, 1993). The same work found that the speaker, not the audience, laughs most, and that laughter is roughly thirty times more frequent in social settings than in solitary ones — a person alone with the same amusing material seldom laughs aloud (Provine, 2000). Juliane Vettin and Dietmar Todt confirmed in detailed recordings of conversation that laughter is frequent, brief, and often produced by the person who is speaking, and that its acoustic features vary with the conversational context in which it occurs (Vettin & Todt, 2004).

As a signal, laughter coordinates and advertises social relationships. Gregory Bryant and a large collaboration played brief clips of colaughter — two people laughing together — to listeners in twenty-four societies around the world, and found that listeners everywhere could judge, above chance and from the sound alone, whether the laughers were friends or strangers, because the spontaneous colaughter of friends has acoustic properties that mark a genuine bond (Bryant et al., 2016). Sophie Scott and colleagues drew these threads together in arguing that laughter is best understood as a social emotion and a communicative act rather than a simple readout of mirth: it manages interactions, signals affiliation and inclusion, and is exchanged strategically as well as spontaneously (Scott et al., 2014). The second demonstration lets the reader explore how the probability of laughter rises with social company and falls in solitude.

Laughter is far more a social act than a private one. Choose a setting and a role and compare the relative rate of laughter, taking laughing alone as the baseline of one.

Note. Rates are relative to laughing alone, set at one. The roughly thirty-fold social increase and the speaker’s advantage over the audience follow Provine’s naturalistic observations; the exact multipliers here are illustrative, built from fixed values (Provine, 1993; Provine, 2000).

The Brain and Evolution of Laughter

Laughter's two forms rest on two neural routes. Barbara Wild, Willibald Ruch, and colleagues reviewed the evidence that spontaneous, emotional laughter is driven by an involuntary system running through the brainstem and limbic structures, while volitional laughter recruits a voluntary premotor and cortical pathway — the same dissociation that produces pathological laughter, the uncontrollable laughing seen after certain strokes and in some neurological disease, where the emotional route is released from cortical control (Wild et al., 2003). On the perceptual side, Carolyn McGettigan, Scott, and colleagues found that hearing volitional laughter engages brain regions for mentalizing — inferring another person's intentions — more strongly than spontaneous laughter does, because a deliberate laugh poses the question of what the laugher means by it (McGettigan et al., 2015). That amusement and its vocalization engage reward circuitry is consistent with imaging of humor, which modulates the mesolimbic reward centers of the brain (Mobbs et al., 2003).

Why laughter should bind people together has a physiological answer. Robin Dunbar's group showed that bouts of social laughter raise the pain threshold — a standard proxy for endorphin activity — more than equivalent non-laughing activity, suggesting that laughter triggers the endorphin release that underpins social bonding in primates (Dunbar et al., 2012). Sandra Manninen, Lauri Nummenmaa, Dunbar, and colleagues confirmed the mechanism directly with positron-emission tomography, showing that social laughter triggers endogenous opioid release in the human brain, and that the density of a person's opioid receptors predicts how much they laugh (Manninen et al., 2017). These functions have deep roots. Jaak Panksepp and Jeffrey Burgdorf discovered that young rats emit high-frequency ultrasonic chirps during play and tickling, and argued that these calls are an evolutionary antecedent of human laughter and joy (Panksepp & Burgdorf, 2003). Marina Davila-Ross and colleagues extended the lineage to our nearest relatives, reconstructing the evolution of laughter from the tickle-induced vocalizations of the great apes and placing human laughter on a continuous phylogenetic tree with theirs (Davila-Ross et al., 2009). Matthew Gervais and David Sloan Wilson synthesized this comparative and functional evidence into an evolutionary account in which spontaneous Duchenne laughter evolved first, as an honest signal of play and safety, and the deliberate non-Duchenne laugh was later recruited for strategic social use (Gervais & Wilson, 2005). The third demonstration turns to how reliably listeners can tell these two kinds of laughter apart.

Drag the effect size d — how far apart listeners’ authenticity ratings of spontaneous and volitional laughter fall — and watch it become the plain-language chance that a random genuine laugh is rated more authentic than a random posed one.

Note. At d = 0 the two kinds are indistinguishable and the chance is 50%. A representative discrimination of d = 1.0 gives about 76%; very clear helpless laughter near d = 1.6 reaches about 87%; a convincing social laugh can pull the separation down to d = 0.4, only about 61%. The conversion is exact for two equal-variance normal distributions; the effect sizes are illustrative (Lavan et al., 2016; Bryant et al., 2016).

Table 1

Spontaneous Versus Volitional Laughter

Note. The two kinds are not sharply separate in practice — a conversation blends them — but they differ reliably enough in production and perception to be told apart (Lavan et al., 2016; McGettigan et al., 2015).

Worked Example

Studies of laughter perception ask whether listeners can tell spontaneous laughter from volitional laughter, and report the result as a standardized effect size, Cohen's d, the separation between the two distributions of listener responses in units of standard deviation. Discrimination of authentic from posed laughter is good but not perfect; take a representative value of d = 1.0. A number like 1.0 is hard to feel, so it helps to convert it into something concrete: the common-language effect size (CLES), here the probability that a randomly chosen spontaneous laugh is rated more authentic than a randomly chosen volitional one.

For two normal distributions with equal variance, that probability is Φ(d / √2), where Φ is the standard-normal cumulative distribution. With d = 1.0, the difference is scaled by √2 ≈ 1.414, giving 1.0 / 1.414 ≈ 0.707. Reading the standard-normal table at 0.707 gives Φ(0.707) ≈ 0.760. So a discrimination of d = 1.0 means that about 76% of the time, a randomly chosen genuine laugh sounds more authentic than a randomly chosen deliberate one — reliable, clearly better than the 50% of a coin flip, but far from the perfect telling-apart implied by the intuition that a fake laugh is always audible. The same formula makes the range vivid. Very clear cases of helpless, spontaneous laughter might reach d = 1.6, giving Φ(1.6 / 1.414) = Φ(1.131) ≈ 0.871, roughly an 87% advantage; a skilled, convincing social laugh might pull the separation down to d = 0.4, giving Φ(0.283) ≈ 0.611, only a little better than chance. The representative figures here are illustrative rather than exact estimates from any one study, but the conversion is exact, and it captures why the authenticity of laughter is audible yet imperfectly so (Lavan et al., 2016; Bryant et al., 2016). The third demonstration implements this conversion.

Discussion

The three questions that open this article have converging answers. A laugh is a stereotyped vocal bout, built on a distinctive respiratory and facial pattern, that comes in two kinds — a spontaneous, emotionally driven laugh and a deliberate, volitional one — differing in acoustics, in neural origin, and in how they are heard. Laughter is, above all, a social signal: it punctuates speech, is produced far more in company than alone, spreads between people, and carries honest information about relationships. And it is old, built on mammalian play vocalizations and tied to the opioid system that supports primate social bonding. What ties the three together is the recognition that laughter is not primarily a response to jokes but an instrument of social life, much of it operating below the level of deliberate control.

This reframing is what distinguishes the cognitive-psychological study of laughter from the folk view that laughter is simply what humor produces. The humor-centered picture gets the direction of explanation backward: laughter is the more basic and more ancient phenomenon, present in play before any joke, and humor is one powerful but not necessary route to it. The spontaneous-volitional distinction, meanwhile, dissolves the apparent paradox that laughter can be both helplessly genuine and socially strategic — these are two systems, one emotional and involuntary, one voluntary and speech-like, that happen to produce overlapping sounds. Seen this way, laughter joins crying, smiling, and facial expression as a channel of nonverbal communication whose meaning lies in what it does between people rather than in what it reveals about a private inner state.

Current Directions

The most active line of work concerns the perception of authenticity and what it demands of the listener's brain. The finding that volitional laughter recruits mentalizing systems more than spontaneous laughter does has opened a program on how listeners decode the intent behind a laugh, how this develops, and how it varies between people (McGettigan et al., 2015). The acoustic basis of the spontaneous-volitional distinction is being mapped in increasing detail, with the aim of identifying which cues — pitch variability, voicing, duration, breathiness — carry the signal of authenticity that listeners exploit across cultures (Lavan et al., 2016; Bryant et al., 2016).

A second strand pursues the neurochemistry of social bonding. The demonstration that shared laughter releases endogenous opioids has prompted questions about how laughter compares with other bonding behaviors such as music, dance, and grooming, whether its effect scales with group size, and how individual differences in the opioid system shape sociability and its disorders (Dunbar et al., 2012; Manninen et al., 2017). Both strands move the study of laughter away from treating it as a byproduct of humor and toward explaining it as a specialized system for managing social relationships, with its own acoustics, its own neural routes, and its own evolutionary history.

Common Misconceptions

Laughter is a response to jokes and humor.

Most everyday laughter follows ordinary conversational remarks, not jokes, and the speaker laughs more than the listener. Laughter is a social vocalization that long predates humor in evolution and in development, and humor is only one route to it (Provine, 2000; Gervais & Wilson, 2005).

A laugh is a laugh; all laughter is the same thing.

Spontaneous, emotional laughter and deliberate, volitional laughter are produced by partly separate brain systems, differ in acoustic structure, and are perceived differently; the dissociation is stark in pathological laughter, where the emotional route is released from voluntary control (Wild et al., 2003; McGettigan et al., 2015).

Laughter is uniquely human.

Tickle- and play-induced vocalizations that are plausibly homologous with laughter occur in the great apes and, in ultrasonic form, in rats, placing human laughter on a continuous evolutionary lineage of mammalian play signals (Panksepp & Burgdorf, 2003; Davila-Ross et al., 2009).

Glossary

Bout.

A single episode of laughter, consisting of a series of short vowel-like notes produced on one exhalation and typically lasting a few seconds.

Colaughter.

Laughter produced by two or more people at the same time; its acoustic properties carry information about the relationship between the laughers.

Contagion.

The tendency of laughter to spread from one person to others who hear it, often without any shared amusing stimulus.

Duchenne Laughter.

Spontaneous, emotionally driven laughter, named by analogy with the genuine Duchenne smile; the honest signal of amusement or play.

Endorphin.

An endogenous opioid peptide released during social laughter; its activity, indexed by raised pain thresholds, is thought to support social bonding.

Mentalizing.

Inferring another person's mental states and intentions; hearing volitional laughter recruits the brain's mentalizing systems more than spontaneous laughter does.

Mirth.

The positive emotional state of amusement that often, but not necessarily, gives rise to spontaneous laughter.

Non-Duchenne Laughter.

Deliberate, volitional laughter produced for social purposes rather than driven by felt amusement; recruited later in evolution for strategic use.

Pathological Laughter.

Uncontrollable laughter, often divorced from felt emotion, caused by neurological damage that releases the involuntary laughter pathway from cortical control.

Play Vocalization.

A call emitted by a young mammal during play or tickling; the proposed evolutionary antecedent of human laughter, seen in apes and in the ultrasonic chirps of rats.

Punctuation Effect.

Provine's finding that laughter falls at the natural breaks in speech — the ends of phrases and sentences — rather than interrupting its flow.

Spontaneous Laughter.

Emotionally driven, largely involuntary laughter generated by a subcortical and limbic pathway; higher and more variable in pitch and less speech-like than volitional laughter.

Tickling.

Physical stimulation that reliably induces laughter in humans and play vocalizations in other mammals, used experimentally to study laughter across species.

Volitional Laughter.

Laughter produced deliberately and under voluntary control, generated by a cortical and premotor pathway and carrying acoustic features of the speaking voice.

Key Researchers

Gregory A. Bryant

. Professor in the Department of Communication at the University of California, Los Angeles; he studies vocal emotion and the cross-cultural perception of spontaneous versus volitional laughter, including colaughter across twenty-four societies. Wikidata - Google Scholar - Faculty Page

Marina Davila-Ross

. Associate professor in comparative psychology at the University of Portsmouth; her work reconstructs the evolution of laughter and play vocalizations across the great apes. ORCID - Faculty Page

Robin I. M. Dunbar

. Emeritus professor of evolutionary psychology at the University of Oxford; his research on social bonding and the social-brain hypothesis established the endorphin-releasing role of shared laughter. Wikipedia - Wikidata - Faculty Page

Carolyn McGettigan

. Chair in speech and hearing sciences at University College London and head of the Vocal Communication Laboratory; she studies the neural basis of voice and laughter perception, including the role of mentalizing in judging authenticity. Google Scholar - Faculty Page

Jaak Panksepp

(1943-2017). Affective neuroscientist who coined the term; his discovery of play- and tickle-induced ultrasonic vocalizations in rats proposed an evolutionary antecedent of human laughter and joy. Wikipedia - Faculty Page

Robert R. Provine

(1943-2019). Neuroscientist and psychologist at the University of Maryland, Baltimore County; his naturalistic study of laughter established the punctuation effect and laughter's primarily social character. Wikidata - Faculty Page

Willibald Ruch

. Emeritus professor of personality psychology at the University of Zurich; he described the facial and respiratory pattern of laughter and studies cheerfulness, exhilaration, and gelotophobia. ORCID - Wikidata - Faculty Page

Sophie K. Scott

. Professor of cognitive neuroscience and director of the Institute of Cognitive Neuroscience at University College London; her research on voice and speech reframed laughter as a social emotion and communicative act. ORCID - Wikipedia - Google Scholar

Frequently Asked Questions

What is laughter?

Laughter is a stereotyped vocalization made of a rapid series of short, vowel-like notes produced on one exhalation, riding on a distinctive respiratory and facial pattern. Cognitive psychology treats it first as a social signal rather than as a simple response to humor, and it is produced largely without deliberate control (Ruch & Ekman, 2001; Provine, 2000).

Is laughter really about jokes?

Mostly not. Naturalistic recording shows that the great majority of everyday laughter follows ordinary conversational remarks rather than jokes, and that speakers laugh more than their listeners. Laughter is a social instrument that humor can trigger but does not define (Provine, 2000).

What is the difference between real and fake laughter?

Spontaneous laughter is emotionally driven and largely involuntary, produced by a subcortical pathway; volitional laughter is produced on purpose by a cortical, speech-like pathway. The two differ in pitch, variability, and duration, and listeners can tell them apart above chance (Lavan et al., 2016; Wild et al., 2003).

Can people tell a genuine laugh from a posed one?

Yes, reliably but not perfectly. Across cultures, listeners judge authenticity from sound alone better than chance, and hearing a deliberate laugh engages brain systems for inferring intent more than a genuine laugh does (Bryant et al., 2016; McGettigan et al., 2015).

Why is laughter contagious and social?

Laughter is produced far more in company than in solitude and spreads between people. As a signal it coordinates conversation, advertises affiliation, and binds groups, which is why it is roughly thirty times more frequent with others than alone (Provine, 2000; Scott et al., 2014).

Does laughing together actually bond people?

There is direct evidence that it does. Shared laughter raises pain thresholds, a marker of endorphin activity, and brain imaging shows that social laughter releases endogenous opioids, the same system that supports social bonding in primates (Dunbar et al., 2012; Manninen et al., 2017).

Do animals laugh?

Tickle- and play-induced vocalizations that appear homologous with laughter occur in the great apes, and young rats emit ultrasonic chirps during play. Human laughter sits on a continuous evolutionary lineage of mammalian play signals (Panksepp & Burgdorf, 2003; Davila-Ross et al., 2009).

What is pathological laughter?

It is uncontrollable laughing, often disconnected from felt emotion, caused by neurological damage that releases the involuntary laughter pathway from the cortical control that normally restrains it. It shows that emotional and voluntary laughter depend on separable brain systems (Wild et al., 2003).

References