A Pompous Snack: On the Unreasonable Complexity of the World's Third-Worst Jokes

Although studies of humour are as old as the Western academic tradition, most theories are too vague to allow for modelling and prediction of humour judgments. Previous work in modelling humour judgments has succeeded by focusing on the world's worst jokes: the slight humour of single nonwords (Westbury, Shaoul, Moroschan, & Ramscar, 2016) and single words (Westbury & Hollis, 2019). Here that work is extended to the world's third-worst jokes, adjective-noun pairs such as dancing dildo, flabby goldfish, and pompous snack. Participants used best-worst scaling to rate the humour of random word pairs. Those judgments were modelled using both linear regression and genetic programming, which is not constrained by assumptions of linearity. The linear regression models were as successful as the nonlinear models at predicting humour judgments, accounting for 27% of the variance in a 540-item validation set. Predictors associated only with the noun and with the relationship between the adjective and noun accounted for much more variance (over 14% each) than predictors associated only with the adjective (6.3%). Greater cosine distance of the adjective word2vec vector from the vectors of the shared neighbors of the noun and adjective is associated with higher humour ratings, whereas the opposite relationship is true for the noun. This captures a form of incongruity not seen in single items, by which neighbours of the adjective become unexpectedly relevant only when the noun brings them into focus. Bien que les études sur l'humour remontent à aussi loin que la tradition universitaire occidentale, la plupart des théories sont trop vagues pour permettre de modéliser et de prédire les jugements relatifs à l'humour. Les travaux antérieurs visant à modéliser les jugements relatifs à l'humour ont connu du succès en centrant leur attention sur les pires blagues au monde : l'humour (discret) des non-mots (Westbury, Shaoul, Moroschan et Ramscar, 2016) et des mots uniques (Westbury et Hollis, 2019). Ici, ces travaux sont étoffés pour inclure les troisièmes pires blagues au monde, soit les paires nom-adjectif, par exemple : dildo dansant, poisson rouge flasque, grignotine hypocrite. Les participants devaient utiliser une échelle allant de la meilleure blague à la pire blague pour évaluer la qualité humoristique de paires de mots aléatoires. Ces jugements étaient ensuite modélisés en utilisant les méthodes de la régression linéaire et de la programmation génétique, laquelle n'est