Investigating Irony Comprehension in Children: Methods, Challenges, and Ways Forward

We review existing developmental research which suggests that verbal irony comprehension is a relatively late acquisition, emerging around the age of five to six years and developing further into adolescence. While children’s difficulty with verbal irony has typically been explained as a result of the complexity of the phenomenon—in particular, that it requires an ability for second-order metarepresentation—it is still puzzling why irony should be so challenging to children given their early sensitivity to other types of language use (e.g., humor, jokes, pretense) that share some of irony’s features (e.g., salient tone of voice, dissociation from explicit/literal utterance content). We discuss some methodological challenges that arise in experimental research on verbal irony comprehension with developmental populations, in particular, differences in operational definitions of irony as well as the types and properties of tasks used to tap irony comprehension in children. Furthermore, we discuss how variability in the extent to which children are exposed to irony, as well as expectations raised by the experimental setting itself, make this non-literal device particularly challenging to investigate in laboratory experiments. In previous work, we have proposed using “implicit” irony comprehension tasks (e.g., tasks that do not require any verbal responding) as a promising new direction for gaining better, more accurate knowledge of children’s developing competence with verbal irony. Here, we present the results of such a task called “the irony game,” used with 91 three- to seven-year-old children. Children had to infer the content of a box, where the correct choice depended on their appropriate literal or ironic understanding of the speaker’s utterance. Only at six or seven years, some children managed to select the correct object in the irony condition of this task. We discuss, from a methodological point of view, the challenges and advantages of our task and suggest ways to improve irony tasks in future research.