Modeling Satirical Uptake Using Discourse Processing Methods

Informed by a theoretical model of satirical uptake, this study investigated processing behavior and comprehension of satirical news articles. Reading times for segments of minimally different satirical and non-satirical texts were collected using within-subjects (Experiment 1) and between-subjects (Experiment 2) designs. Segment reading times and participant familiarity with news genres were used to predict ratings of sincerity, humor, and agreement, as well as manually coded comprehension scores for the satirical texts. In both experiments, text perceptions were significantly different for satirical (vs. non-satirical) texts, with some processing differences observed in Experiment 1. Further results from Experiment 1 included no effects for segment reading times on text perceptions or comprehension scores but did include effects for genre familiarity on text perceptions. Experiment 2 results indicated slower reading times were associated with higher perceptions of sincerity and lower chances of satire comprehension, suggesting effortful processing is a marker of failed satirical uptake.