A re-visit of three-stage humor processing with readers' surprise, comprehension, and funniness ratings: An ERP study

- The roles of surprise, comprehension, and amusement levels in humor processing were examined.
- Participants were divided into high/low score groups based on their behavioral ratings to verbal jokes.
- Highly surprised, comprehended, and amused group elicited larger N400, P600, and LPP effects, respectively.
- These intergroup variances supported the three-stage model of humor processing.

Humor processing can be divided into three sub-stages including incongruity detection, incongruity resolution, and elaboration (Chan et al., 2012, 2013; Feng et al., 2014). However, few studies have investigated the three-stage model of humor processing with readers' surprise, comprehensibility and funniness levels, and little discussion has been devoted to its biological underpinning. To verify the credibility of the three-stage model, electroencephalography (EEG) was utilized in corroboration with two types of stimuli including jokes and non-jokes in the present research. Participants were categorized into high vs. low score groups based on their rating scores of surprise, comprehension, and funniness to joke stimuli. The between-group analyses showed that compared with the less surprised group, highly surprised people elicited a primarily larger N400, which may suggest more incongruity perceived in reading jokes. Additionally, good comprehenders mainly elicited a larger P600, probably indicating a more successful resolution of detected incongruity in comparison with poor comprehenders. Finally, the highly amused group elicited a larger late positive potential (LPP) compared with the less amused group, which could reflect more affective elaboration of jokes. Participants' surprise, comprehension, and funniness levels had smaller impacts on other chief electrophysiological components, with the effects varying with different group contrasts. These results provided the evidence that different degrees of surprise, comprehensibility, and amusement to jokes would influence the three sub-stages (incongruity detection, incongruity resolution, and elaboration) respectively in humor processing. The current study thus generally re-verified the stability of the three-stage model through participants' behavioral ratings which had seldom been touched upon.