Attentional biases toward humor: Separate effects of incongruity detection and resolution

Previous research has indicated that the comprehension of humor involves two stages: incongruity detection and incongruity resolution. However, little is known about the temporal parameters of these stages and the degree to which they influence attentional processing. In the current study, 155 participants completed a dot-probe task to examine these questions. On each trial, humor versus control, novel (incongruent but not humorous) versus control, or neutral versus neutral image pairs were presented for 300, 400, or 500 ms. A probe immediately replaced the experimental image (valid trial) or control image (invalid trial). An attentional bias toward humor and novelty by 300 ms was shown by faster probe-detection reaction times (RTs) on valid compared to invalid humor and novel trials for all three exposure times. When compared to the neutral trials, humor and novel stimuli elicited slower RTs, indicating a difficulty in attentional disengagement. An exploratory analysis found that subjective humor ratings predicted the disengagement bias at 500 ms, but not at 400 or 300 ms. These results suggest that incongruity detection biases attention by 300 ms, whereas incongruity resolution may only contribute at 500 ms.