A University of Alberta researcher has scientifically determined that “upchuck” may be the funniest word in the English language.
U of A psychologist Chris Westbury co-authored a study in the Journal of Experimental Psychology: General in October on why words like “wriggly,” “squiffy,” and “boob” were funny.
Westbury said he got the idea for this study when he noticed aphasia patients (who suffer from brain-injury-related language impairment) would often laugh at gibberish words, particularly “snunkoople.” Studying these non-words, his team determined that the humour quotient of made-up words could be predicted by the average frequency of the letters in them.
Later, he was a reviewer of a paper by the University of Warwick’s Thomas Hill, who had people rank 4,997 words based on their funniness. Realizing that Hill’s team did not investigate why those words were funny, Westbury decided to do his own analysis.
Westbury’s team developed a statistical model based on the Hill sample to see if it could figure out what made words funny. The model studied the form of the words (letter and sound probabilities) and their semantics (relation to emotions and six categories of funny words: insults, sex, partying, animals, bodily functions and expletives).
The team found that the best predictor of a word’s funniness was its average distance from the six subject categories. Westbury said this made sense as funny words like “boobs” often fell across several categories (sex and bodily functions).
Certain letters and structures also made words more funny, the model found. As many comedians have noted, the letter “k” was statistically more likely to occur in words deemed funny than those that were not – much to Westbury’s surprise. The “oo” sound in “boot” and words that ended in “(consonant)-le” or “y” were also more likely to show up in silly words (such as “moo,” “chuckle,” and “silly”).
Applying this model to a list of 45,516 words, the team predicted that the 10 funniest were (in descending order) upchuck, bubby, boff, wriggly, yaps, giggle, cooch, guffaw, puffball, and jiggly. The least funny was harassment. The team then had 74 U of A students rank groups of words as funny or unfunny. The human rankings were pretty close to those predicted by the model, suggesting it was accurate.
Westbury said this was one of the first studies to produce a model that could reliably predict whether or not something was funny. More importantly, it shows that the two leading humour theories (superiority, or “we laugh when someone is made to seem inferior to us,” and incongruity, or “we laugh when expectations are violated”) are way too simplistic, as so many elements appear to affect funniness.
“One of the weird things about humour is that we laugh at things that have no similarity to each other at all,” Westbury said. If an alien asked us why a limerick and a man slipping on a banana peel were both funny, we could not answer.
Hill, reached by email, said this study was fascinating due to the amazing number of factors it found that contributed to single-word humour.
“If we imagine scaling that up to jokes, we begin to understand why humour is so difficult to do well.”
Humour is important as it expresses a lot of information about social groups, intelligence and perspective, and “may be the last bastion of humanity against artificial intelligence,” Hill said. Understanding it can help us make predictions about aging, memory, language acquisition and other mental processes.
“It may seem ridiculous, but humour lurks at the heart of many psychological disciplines.”
Westbury said he doubted humans would ever figure out some sort of grand unified humour theory, as it seems to be determined by the sum total of human experience.
“The grand theory of humour really is ‘a human being,’ ” he said.
