Poster #113 - Scalar implicature derivation in 4- and 5-year-olds is supported by language and executive function networks

• In Event: 3-159 - Poster Session 12
  In Poster Session: PS 12 Section - Language, Communication

Integrative Statement

Conversational principles lead adults to interpret a sentence like “Some children eat broccoli” to mean not all children do (Grice, 1989). This inference, a “scalar implicature” (SI), requires pragmatic reasoning about speaker intentions, which relies on complex linguistic and cognitive abilities. Neuroimaging studies show recruitment of cognitive (i.e. executive functions; EF) and social (i.e. Theory of Mind; ToM) networks during pragmatic reasoning in adults (van Ackeren et al., 2012; Champagne-Lavau & Stip, 2010); however, little is known about the neural systems underlying development of pragmatic and social reasoning before age 5 (Shetreet et al., 2014; Gweon et al., 2012). Pragmatic reasoning develops during preschool, with some SI studies reporting success as early as 3.5-years (Stiller et al., 2015), when cognitive (EF) and social (ToM) abilities are concurrently developing. However, other studies have reported SI failures in children as old as 9 (Noveck, 2001). Examining developing brain networks can provide insight into the trajectories of pragmatic development to inform current inconsistencies that behavioral results alone cannot address. We ask how cognitive, linguistic, and social abilities, supported by the maturation of specific neural systems, contribute to pragmatic development. Using functional Near Infrared Spectroscopy (fNIRS) neuroimaging, we investigate how neural systems that support SI development change between ages 4 and 5, and into adulthood.

Participants completed a behavioral battery (PPVT, DCCS, Sally-Anne task) and neuroimaging SI task (adapted from Kampa & Papafragou, 2017). In this referent selection task, participants saw a pair of photos (Fig.1) which differed only in whether the observer had full visual access to the contents of the box (full-knowledge) or limited visual access (limited-knowledge). Participants were told that the girl would describe what she sees in one of the two boxes. Participants would hear either “I see a spoon and a bowl” (strong) or “I see a spoon” (weak); a pragmatic responder should link the strong statement to the full-knowledge speaker and the weak statement to the limited-knowledge speaker, in accordance with principles of informativeness. Imaging data were analyzed using NIRS-SPM-v4 (Jang et al., 2009).

Results indicate activation of linguistic and cognitive networks during SI derivation for children and adults (Fig.2). Adults (N=17) showed greater activation for the weak vs. strong condition (i.e. implicature derivation) in the left VLPFC, including Inferior Frontal Gyrus (LIFG), a region associated with higher-level linguistic and cognitive processing, and the right DLPFC, associated with EF. 4- and 5-year-old children (initial pilot: N=7) who succeeded on the task displayed similar patterns of activation in both regions, suggesting early recruitment of language and EF networks in pragmatic reasoning. Ongoing work will yield further insight into neural recruitment for SI in children with varying pragmatic reasoning ability.

These data provide new insight into the pragmatic development in 4- and 5-year-olds; by age 4, some children show adult-like recruitment of language and executive function networks during scalar implicature derivation. These findings not only address age inconsistencies in the behavioral literature, but also provide broader insights into the development of linguistic and cognitive neural systems in young children.

Authors

• Alyssa Kampa, University of Delaware
  Presenting Author

• Kaja K Jasinska, University of Delaware
  Non-Presenting Author

• Anna Papafragou, University of Delaware
  Non-Presenting Author