Event end-state encoding in 13-month-olds—completed and non-completed events are different

Abstract

Young children sometimes incorrectly interpret verbs that have a “result” meaning, such as understanding ‘fill’ to refer to adding liquid to a cup rather than filling it to a particular level. Given cross-linguistic differences in how event components are realized in language, past research has attributed such errors to non-adultlike event-language mappings. In the current study, we explore whether these errors have a non-linguistic origin. That is, when children view an event, is their encoding of the event end-state too imprecise to discriminate between events that do versus do not arrive at their intended endpoints? Using a habituation paradigm, we tested whether 13-month-old English-learning infants (N = 86) discriminated events with different degrees of completion (e.g., draw a complete triangle versus draw most of a triangle). Results indicated successful discrimination, suggesting that sensitivity to the precise event end-state is already in place in early infancy. Thus, our results rule out one possible explanation for children’s errors with change-of-state predicates—that they do not notice the difference between end-states that vary in completion.

Boundedness supports children’s event representations.

“…As alluded to earlier, the literature on event cognition typically assumes that event boundaries have a privileged status in memory and provide anchors for later learning and describing (Swallow et al, 2009). On this view, event endpoints, in particular, are critical for how events are represented by both children and adults (see He & Arunachalam, 2023; Lakusta & Landau, 2005, 2012; Papafragou, 2010; Regier & Zheng, 2007; Strickland & Keil, 2011). This literature has mostly drawn its examples from events with self-evident endings.…”