In the studies of aesthetic appreciation, poetry has received less attention than other art forms.Furthermore, although a number of recent studies have focused on the factors that influence poetry appreciation, very few attempts have been made to explain the appreciation process.In this study, to explore the cognitive process of poetry appreciation, we propose a computational model of poetry appreciation based on the incongruity resolution theory, which has been proposed to explain the process of evoking poetic effects.Assuming that poetry reading comprehension is a process of evoking mental imagery from words in a poem, we model it as spreading activation in a semantic network.The notion of incongruity resolution is then modeled as the difference of the diversity of activation in the network. The computational model was tested using multiple regression analysis with simulation results as independent variables and human ratings for aesthetic appeal as a dependent variable. The analysis demonstrated that the degree of incongruity resolution computed by the model accounted for a significant portion of the variance in aesthetic appeal. This finding provides evidence for the validity of the model and suggests its potential for a more comprehensive model of poetic appreciation in literary works.