PUBLICATIONS

The usage-based constructionist approach to language offers a clear perspective and suggests testable ways to mitigate the special challenges language learning presents to autistic individuals.

Publications

Conventional metaphors (e.g., a firm grasp on an idea) are extremely common. We have discovered a new reason they are used so often: they are more engaging in that they evoke more focused attention than carefully matched literal paraphrases (e.g., a good understanding of an idea) or concrete descriptions (e.g., a firm grip on a doorknob). We initially discovered this accidentally by observing in an fMRI study that the amygdala—standardly used to indicate heightened engagement to tasks—was more active when participants passively read metaphorical sentences in the fMRI than when they read literal paraphrases that differed only by a single word (Citron & Goldberg, 2014). Since then, we’ve probed this discovery in a number of ways and have confirmed the effect, regardless of which conventional metaphors are used or whether they are included in longer passages or isolated sentences (Citron et al., 2016; Citron, Michaelis & Goldberg, 2020). In each study, sentences were normed and matched on complexity, plausibility, emotional valence, intensity, familiarity. Most recently, we measured pupil dilation in a light-controlled room as sentences were read aloud.  As predicted, metaphorical sentences elicited greater event-evoked pupil dilation compared to literal paraphrases or concrete descriptions, beginning immediately and lasting seconds beyond the end of the sentences. We conclude that conventional metaphors are more engaging than literal paraphrases or concrete sentences in a way that is irreducible to difficulty or ease, amount of information, short-term lexical access, or downstream inferences.

PUBLICATIONS

Parallels between the usage-based constructionist perspective and LLMs