The nascent lexicon: early word learning in the lab and in the world

While a longstanding view in language development holds that infants don't understand words until they begin talking (around age 1), recent research from eyetracking studies in the lab has revealed that infants begin understanding words months earlier (e.g., Bergelson & Swingley, 2012; 2015; Tincoff & Jusczyk 1999; 2012; Parise & Csibra, 2012; Bergelson & Aslin 2017; Kartushina & Mayer, 2019). In this talk I will explore two branches of my lab’s work that begin to unpack the mechanisms of early word learning: studying the learner, and studying the learning environment. In this talk I will discuss eyetracking data revealing infants’ initially immature expectations about how words sound and what they mean, and how their representations eventually become more adult-like over infancy and toddlerhood as early phonology, semantics, syntax, and morphology come online. Synthesizing across studies, I will discuss recent results showing a robust, non-linear, and arguably qualitative improvement in infants’ real-time word comprehension just after the first birthday. Drawing from SEEDLingS, my lab’s audio and video corpus of home recordings, I will argue that this “comprehension boost” is not well-explained by changes in language input for common words, but rather, by postulating that infants learn to take better advantage of relatively stable input data. I will propose complementary theoretical accounts of what makes older infants “better learners.” Time permitting, I will also discuss the dynamics of language learning beyond our typical WEIRD populations, calling on data from cross-cultural collaborations, and early stage work looking at early language development infants who are blind and infants who are Deaf/Hard of Hearing.

Contributions to and/or sponsorship of any event does not constitute departmental or institutional endorsement of the specific program, speakers, or views presented.