No Need to Laugh Out Loud: Predicting Humor Appraisal of Comic Strips Based on Physiological ...
summary
25 participants browse a live humorous website while wearing lightweight sensors (22 usable); EEG with a single posterior channel, finger EDA, and 3-lead ECG record responses while each comic is labeled funny/not funny. Random forest models reach AUC ≈ 0.73 using fused physiology, ECG alone ≈ 0.55, and combined physiology+video ≈ 0.88. The most predictive features come from a special window (−2s to +5s around joke end) and later trial stages, notably gamma-band EEG and late EDA rises; movement and muscle artifacts remain a caveat.
AI-written summary · about
chapters (8)
- 0:00Introduction to humor prediction
- 1:32Motivations behind research
- 4:19Experiment design overview
- 7:08Predictive modeling approach
- 9:35Key results and findings
- 13:56Camera-based comparison
- 15:42Summary and takeaways
- 17:31Q&A session
related talks
33mWhat if a Robot Could Laugh at Your Jokes? (Laughter Prediction with Natural Language Processing)
33mWhen Stand-up Comedy Meets Machine Learning
35mTranslating Brain Electrical Signals into Actions through Models...
07mAssessing Cognitive Performance Using Physiological and Facial Features: Generalizing Across ...
01h 01mKeynote: Using EEG and Machine Learning to Perform Lie Detection
01h 05mUsing EEG and Azure Machine Learning to Perform Lie Detection
06m"Let’s Get Physiological, Physiological!": A Systematic Review of Affective Gaming
21mFramework for Electroencephalography-based Evaluation of User Experience