How Language Supports the Acquisition of Predicates of Mental States and Emotions

Kristen Syrett, Misha Becker

The Slurring Terms Across Languages (STAL) network (https://sites.google.com/view/stalnetwork/home) invites you to a talk by Kristen Syrett (Rutgers University) & Misha Becker (University of North Carolina - Chapel Hill) entitled "How Language Supports the Acquisition of Predicates of Mental States and Emotions". The talk will take place online on JANUARY 20, 14:30-16:00 Central European Time (CET), and is part of the of STAL Seminar series (https://sites.google.com/view/stalnetwork/seminar). If you want to participate, please write to [email protected] for the Zoom link. Below you can find the abstract.

All welcome!

ABSTRACT:

As children acquire adjectives, they must tackle the challenge that while some properties denoted by these predicates are stable and visually salient (e.g., color, shape), others (e.g., emotions and mental states like happy, sad, or confident) lack a reliable physical correlate, and are typically only inferable via second order characteristics. How, then, do children master the meanings of adjectives that label these fleeting, internal, abstract states? One answer may lie in the very linguistic environment in which these adjectives appear. Previous work in language acquisition has documented the power of the frame and complementation patterns for verb learning, subject form for control and raising verbs, count syntax for acquiring nouns, and adverbial modification for different types of gradable adjectives. In this talk, I draw on this prior work to lay a foundation for a series of experiments investigating how children might recruit both syntactic and semantic cues in the input to narrow the hypothesis space for emotion/mental state adjective meaning. I begin by presenting extensive evidence from CHILDES corpora showing that while these adjectives are relatively infrequent in the input, they diverge from other adjectives (e.g., those of color, shape, size, or multidimensional subjective adjectives) in their preference of syntactic position, their requirements on subject animacy, and their syntactic complementation patterns. Next, I present data from a set of word guessing studies using scripted dialogues that both adults and older children (age 5-8) recruit the type of subject and syntactic complement to constrain adjective meaning. Finally, I present a set of binary forced-choice word learning studies putting emotion/mental state against color and shape showing once again, that the presence of an animate subject and syntactic complement points to an emotion/mental state adjective meaning, this time for preschoolers. Taken together, these experiments—the first to document the combined power of syntax and semantics for acquiring abstract adjective meaning—make connections between emotion/mental state adjectives and mental state verbs in word learning, thereby further demonstrating the potential universality of syntactic bootstrapping, and the role of language itself in focusing young word learners’ attention on mental aspects of the situation that are not readily observable.