Humor appreciation can be predicted with machine learning techniques

Hannes Rosenbusch, Thomas VisserScientific 2023 · DOI

Abstract

Humor research aims to predict amusement, yet traditional psychological theories often struggle with individual-level predictive accuracy. This study evaluates the predictability of humor appreciation using machine learning (XGBoost). We find that while demographic and psychological traits offer minimal predictive gains, models incorporating previous ratings from the same rater and the general popularity of the stimulus can reach the theoretical ceiling of predictability. Successful prediction relies heavily on "rating aggregates" and specific rater-stimulus interactions, such as joke comprehension.

Introduction

Humor is fundamentally subjective, raising the question: What makes things funny? While psychological research has identified various traits associated with humor—such as extraversion or "dispositional cheerfulness"—these findings often rely on statistical associations rather than predictive power. This paper bridges the gap by testing whether machine learning can accurately predict if a specific individual will find a specific joke funny.

Methods

Researchers conducted two large-scale studies (N > 5,000 each) using XGBoost (Extreme Gradient Boosting). This algorithm was chosen for its ability to capture complex, non-linear interactions between variables that standard linear models might miss.

  • Study 1 examined broad psychological constructs (Big Five, sensation seeking, etc.) and demographic data.
  • Study 2 was a pre-registered replication adding specific "rater x stimulus" interactions, such as perceived friendliness and understandability.

The "theoretical ceiling" of prediction was established by measuring the test-retest reliability of participants' ratings. This provides a benchmark for the maximum possible predictive accuracy given the inherent noise in human judgment.

Datasets and Variability

The studies utilized a diverse corpus of 105 humor stimuli, with participants rating 10 stimuli each. The data revealed significant variance both across different jokes and across different raters, which the models leveraged for prediction.

Overview of funniness ratings across stimuli and raters for both studies, showing the variance in mean scores and density distributions.

As shown in the distributions, some stimuli are consistently rated as funnier than others, and some raters are consistently more "generous" with their amusement scores. These "aggregates" proved to be the most powerful features for the ML models.

Key Results

The machine learning models achieved an accuracy (AUC ~0.74) that virtually matched the retest-reliability of the participants. Key findings include:

  • The "Personality" Mirage: Psychological scales alone yielded an accuracy of only ~64%. Traits like extraversion offer negligible gains in predicting a specific instance of amusement.
  • Past Behavior is the Best Predictor: Incorporating "rating aggregates"—the rater's average score for other jokes and the joke's average score from other people—boosted accuracy to ~70.5% in Study 1 and ~76% in Study 2.
  • The Role of Comprehension: The strongest predictor in Study 2 was whether the rater understood the specific joke. This interaction between the rater and the content is more critical than general cognitive traits.
Detailed distribution of funniness ratings within raters (Study 2), highlighting the spread of individual amusement levels on a 1-7 scale.

Discussion and Limitations

The models reached the stability limit of the underlying data ($R^2 \approx 0.35$), suggesting that humor is about as predictable as humanly possible. However, limitations remain. The study focused on "performance humor" (static text/images) and did not account for social context or the relationship between the teller and listener. Future research could leverage LLMs for better feature engineering of joke structures or explore real-world social dynamics to further refine these predictions.

Summary updated

Abstract

Humor research is supposed to predict whether something is funny. According to its theories and observations, amusement should be predictable based on a wide variety of variables. We test the practical value of humor appreciation research in terms of prediction accuracy. We find that machine learning methods (boosted decision trees) can indeed predict humor appreciation with an accuracy close to its theoretical ceiling. However, individual demographic and psychological variables, while replicating previous statistical findings, offer only negligible gains in accuracy. Successful predictions require previous ratings by the same rater, unless highly specific interactions between rater and joke content can be assessed. We discuss implications for humor research, and offer advice for practitioners designing content recommendations engines or entertainment platforms, as well as other research fields aiming to review their practical usefulness.
BibTeX
@inproceedings{rosenbusch2023humor,
  author = {Hannes Rosenbusch and Thomas Visser},
  title = {{Humor appreciation can be predicted with machine learning techniques}},
  booktitle = {Scientific},
  year = {2023},
  doi = {10.1038/s41598-023-45935-1},
}
Figure