The story is told this way. Somewhere in Los Angeles in the middle of the 1950s, a room full of people watched a man perform without saying a word, and laughed.
The laugh was recorded. It was cut away from the thing that caused it, spliced into a loop of tape and locked in a box. For the next twenty years it was played under jokes those people never heard, on programs that did not exist when they laughed, into living rooms they never entered. The programs are still rerun.
Most of the people on that tape are dead now. They are still laughing. Sometimes at really bad jokes.
The Audience Reaction Duplicator
The box was built by Charles Douglass, a Navy-trained electronics engineer who worked in television sound. Its official name was the Audience Reaction Duplicator. The industry called it the Laff Box. Accounts describe it as a cabinet more than two feet tall, padlocked shut and played like an organ. Inside, by the usual count, were 320 laughs on 32 loops of tape, ten to a loop: chuckles, titters, belly laughs, the gasp before a roar.
From the late 1950s to the late 1970s that box and its successors were the audience of most American prime-time comedy. Douglass worked from a garage in the San Fernando Valley under the name Northridge Electronics. For years he had the business almost to himself. He let nobody outside his family see inside the machine, and only a few technicians he trained were allowed to play it.
The laughs came from somewhere. The story most often told is that Douglass took them from the silent pantomime sketches on The Red Skelton Show on CBS, and from a performance Marcel Marceau gave in Los Angeles in 1955 or 1956. A mime was ideal. Nobody spoke over the laughter, so it could be lifted out clean. It is a good story, and nobody has audited it. The provenance of the most-heard audience in television history rests on a recollection. That is fitting for a machine built so that nobody could look inside it.
Every laugh in the box was real. None of them was at the joke it was played under.
The audience was the problem
The usual story is that canned laughter was invented to rescue jokes that did not work. Gina Giotta's institutional history of the device, Sounding Live (2017), tells a colder one. What early-1950s television needed rescuing from was the studio audience. A room full of living people is unpredictable. It laughs late, laughs at the wrong line, falls silent, coughs, and cannot be asked to do it again. Television wanted what the audience gave it, the sense of a live and intimate occasion that set it apart from film and let it into the home. It did not want the people.
The laugh track kept the first and removed the second. What remained was an audience with every property of a crowd except presence, one that could be summoned, shaped, extended and dismissed. The usual first date is The Hank McCune Show in 1950, the first sitcom whose laughter came wholly from a recording. Some accounts put an earlier splice, laughter taken from a 1949 Bing Crosby broadcast, behind it. So 1950 is the first sitcom laugh track by the usual definition, not the first recorded laugh.
Four rooms, one of them empty
“Laugh track” names several different things, and the arguments about it usually go wrong by treating them as one.
| Kind | Where the laughter came from |
|---|---|
| Live | A studio audience in the room while the scene was played, recorded with it. |
| Sweetened | A live audience, corrected afterwards: a laugh lengthened, a weak one swelled, an overlong one cut. |
| Canned | Laughter recorded somewhere else, at some other time, at something else, and laid in during editing. |
| None | No audible audience. The viewer is alone with the show and has to decide for themselves. |
The first three reach the viewer as the same sound, which is the whole point of the second and third. Nobody at home can tell them apart, and nobody was ever meant to.
It works best when it was real
In 2019 Sophie Scott's laboratory at University College London tested the device on the worst material it could find: forty jokes chosen for being bad, the kind known as dad jokes. Each was first rated with no laughter at all. A comedian then recorded each one, and seventy-two adults heard them with laughter laid underneath, some of it posed and some spontaneous. The laughter made the jokes funnier, for autistic and non-autistic adults alike. And it worked better when the laughter was spontaneous, recorded from people who were really laughing, than when it was posed.
That is the strange arithmetic of the box. If the story of its origin is true, Douglass's laughs were not performed. They are the genuine, helpless laughter of an audience watching a mime. That is exactly the kind the laboratory found most persuasive. The laugh track is not a fake reaction. It is a true reaction to a missing cause, an honest laugh detached from what it was honest about and fastened to whatever needed it next.
The dead were not pretending. That is why it works.
What the experiments found
Psychologists have tested the laugh track for nearly forty years. Read one at a time, the studies seem to contradict each other. Read together, they turn out to be measuring different things.
| Study | What was varied | What it found |
|---|---|---|
| Neuendorf & Fennell, 1988 | A laugh track or none | Viewers who heard the laugh track laughed more, and did not rate the material significantly funnier. |
| Lawson, Downing & Cetola, 1998 | Laughter believed live, believed canned, or believed constrained | Laughter moved judgements of funniness less when people knew it was canned, or believed the audience had been told to laugh. People treat laughter as evidence of what other people thought. |
| Platow et al., 2005 | Whether the laughing audience was the listener's own group or another | The title is the finding: “It's not funny if they're laughing.” Laughter from the listener's own group raised laughing, smiling and ratings of funniness. Laughter from an out-group did not. |
| Gillespie, Mulder & Leib, 2016 | Simulated laughter or none, across three field studies | The laugh track made shows seem funnier, and did not make them more enjoyable. It acts as a quick cue that something is a joke, and the same cue gets in the way of being absorbed in the story. The authors call it the laugh-track paradox. |
| Rhodes & Ellithorpe, 2016 | A sitcom scene of reckless driving, with its laugh track or with it edited out (112 participants) | The laugh track changed how readily people thought of reckless driving as acceptable, more so among viewers who identified with the reckless character. That in turn predicted their attitudes and stated intentions about driving that way. |
| Waddell & Bailey, 2019 | Canned laughter added to Scrubs, a show made without it, crossed with positive or negative social-media comments | As later reviews report it, the canned laughter had no effect, direct or indirect, on perceived humor or enjoyment. |
| Cai, Chen, White & Scott, 2019 | Forty bad jokes, rated bare, then heard with spontaneous or posed laughter | Any laughter made the jokes funnier. Spontaneous laughter did more than posed. |
Three findings recur. The laugh track works on behavior and judgement more reliably than on pleasure. It can make you laugh without making you think it was funny, and make you think it was funny without making you enjoy it. And it works only as far as the listener believes in the people laughing. It is weaker when they know the laughter is canned, absent when the laughers are the wrong people, and, in the one study that laid it over a show built without it, nothing at all.
Put plainly, a laugh track is a report about other people. It does not tell you a line is funny. It tells you that a room full of people like you found it funny, and that laughing is what is done here. Rhodes and Ellithorpe's study should be quoted most and is quoted least, because it shows the report carries more than a verdict on a joke. It carries a norm about the behavior the joke is about. The dead are not only telling you what is funny. They are telling you what is acceptable.
The objection is not coercion. It is the secret.
The standard complaint is that the laugh track tells you when to laugh. Kenneth L. Brewer's essay Don't Make Me Laugh! (2018) takes that complaint, and the stronger one that the device makes viewers more receptive to what a show is saying, and finds both overstated. People are not passive in front of the sound, and the experiments above agree. What Brewer does find is deception. It shows in the industry's continued reluctance to say when and how its laughter was made. He proposes moving the argument from blame to disclosure. Whether an episode's laughter was live, sweetened or canned is information a viewer is owed.
That is the argument this department would have made, and it is the method this site runs on. The Colophon publishes who wrote each line of the site. The Royalty Ledger publishes the gaps in its own corpus. The laugh track is the opposite instrument, a census of an audience that declines to be audited, kept in a padlocked box. In Lawson's experiment the laughter lost its force the moment people were told it was canned. Disclosure is the one intervention the research says would work, and it is the one the industry never adopted.
What the ledger can hear
Sitcom Royalty holds seventy years of the form and cannot tell you which of its shows laughed. Its Format column records Multicam, Single-camera, Hybrid or Animated. A multicamera show may have been filmed before a live audience, sweetened afterwards or canned from the start, and the column cannot tell them apart, any more than a viewer could. Filing it as one fact would make the same conflation this page has refused.
The ledger holds one thing about the laugh track, and it is an absence. Among its formal devices, the L4 layer, is Single camera without laugh track. It is defined as a departure: film grammar and no audible audience, in a period when the form's default was neither. Once the silence became the default it stopped being a device. One genealogy was proposed for it, from The Monkees to Malcolm in the Middle. It was retired at ruling R090, because the evidence offered was production interviews about rejecting the multicamera audience, and none of them named The Monkees as the ancestor. The ledger keeps the edge and the sentence that retired it.
The laughter has a box, a padlock and a legend. The silence has a definition and no confirmed first link.
What this page cannot prove
That these particular laughs are the dead's. The Skelton and Marceau origin is the story most often told, and the accounts that tell it call it belief or legend. Nobody has played the loops against the source performances. What needs no legend is the arithmetic. An adult in a 1950s television audience was born before the Second World War ended, and most of them have died since. Whatever room the box's laughter came from, that room has mostly emptied.
Which laughs are in a given episode. Live, sweetened and canned laughter reach the ear as one sound, and the industry has not said which an episode used. That is Brewer's point, and it is why this page names no show as carrying the dead.
What was supplied and is not printed here
This page began as a research summary produced by a search assistant. Most of its citations held up, and some did not. What could not be confirmed was left out rather than smoothed over. It is listed here because a reader who has seen those claims elsewhere deserves to know why they are missing.
- A 2026 meta-analysis of laugh tracks by Logunova and Allen, said to pool 35 manuscripts to an average effect of r = .391. A paper on the subject may exist, but no source this page could open confirmed its authors, its sample or its figure. A pooled effect size is exactly the kind of number that gets repeated because it is precise. Until somebody reads it off the paper, it does not go in.
- “Gianluca Giotta”. The author of Sounding Live is Gina Giotta.
- Neuendorf and Fennell in Communication Studies. In 1988 the journal was the Central States Speech Journal, which later took that name. The summary also omitted their most useful result, that the laugh track made people laugh more without making them rate the material funnier.
- The publication details of Waddell and Bailey, given as Psychology of Popular Media Culture 8(1), 99–107. Later reviews that cite the study confirm it and its null result. The volume and pages were not checked against the journal.
- The summary's closing paragraph, which advertised a research product. That was the supplied text's own laugh track, and it is not reproduced.
If you can supply the meta-analysis, or a recording that settles where the box's laughter came from, it goes in. Write to info@themeansofproduction.press.
Sources
- Giotta, G. (2017). Sounding live: An institutional history of the television laugh track. Journal of Communication Inquiry, 41(4), 331–348. doi:10.1177/0196859917713760
- Neuendorf, K. A., & Fennell, T. (1988). A social facilitation view of the generation of humor and mirth reactions: Effects of a laugh track. Central States Speech Journal, 39(1). doi:10.1080/10510978809363233
- Lawson, T. J., Downing, B., & Cetola, H. (1998). An attributional explanation for the effect of audience laughter on perceived funniness. Basic and Applied Social Psychology, 20(4), 243–249. doi:10.1207/s15324834basp2004_1
- Platow, M. J., Haslam, S. A., et al. (2005). “It's not funny if they're laughing”: Self-categorization, social influence, and responses to canned laughter. Journal of Experimental Social Psychology, 41(5), 542–550. ANU Open Research
- Smith, J. (2005). The frenzy of the audible: Pleasure, authenticity, and recorded laughter. Television & New Media, 6(1), 23–47.
- Bore, I.-L. K. (2011). Laughing together? TV comedy audiences and the laugh track. The Velvet Light Trap, 68, 24–34. Project MUSE
- Rhodes, N., & Ellithorpe, M. E. (2016). Laughing at risk: Sitcom laugh tracks communicate norms for behavior. Media Psychology, 19(3), 359–380. doi:10.1080/15213269.2015.1090908
- Gillespie, B., Mulder, M., & Leib, M. (2016). Who's laughing now? The effect of simulated laughter on consumer enjoyment of television comedies and the laugh-track paradox. Journal of the Association for Consumer Research, 1(4), 592–606. doi:10.1086/688220
- Brewer, K. L. (2018). Don't make me laugh! Morality, ethics, and the laugh track. Studies in American Humor, 4(1), 10–36. Project MUSE
- Cai, Q., Chen, S., White, S. J., & Scott, S. K. (2019). Modulation of humor ratings of bad jokes by other people's laughter. Current Biology, 29(14), R677–R678. Current Biology; UCL release
- Waddell, T. F., & Bailey, E. (2019). Is social television the “anti-laugh track”? Testing the effect of negative comments and canned laughter on comedy reception. Cited here through the reviews that report it; see above.
- Wikipedia — Charles Douglass and Laugh track, for the box, Northridge Electronics, the Skelton and Marceau origin and The Hank McCune Show; Hackaday — Charlie Douglass and the laugh track
- The Sitcom Royalty ledger: the
Formatcolumn ofshows.csv; the L4 device Single camera without laugh track; formal edge FE014 and ruling R090. Their state is published on the Royalty Ledger.