Discovering Abstract Structures in Infancy

How do infants learn about the formal properties of language using only cues they can access in speech? And what intuitions do they bring to the learning problem? Chapter 2: To explore whether current notions of statistically-based language learning could successfully scale to infants’ linguistic experiences “in the wild”, we implemented a statistical-clustering word-segmentation model (Saffran et al., 1996) and sent its outputs to an implementation of a “frame” based form-class tagger (Mintz, 2003) and, separately, to a simple word-order heuristic parser (Gervain et al., 2008). We tested this pipeline model on various input types, ranging from quite idealized (orthographic words) to more naturalistic resyllabified corpora. Chapter 3: When infants hear sentences containing unfamiliar words, are some language-world links (such as noun--object) more readily formed than others (verb--predicate)? What if the context renders verb-predicate and noun-object interpretations equally plausible? We examined 14-15-month-olds’ capacity for linking semantic elements of scenes with simple bisyllabic nonce utterances using 2AFC language-guided looking. Chapter 4: One important function of phonology is to set rules about what phonetic strings count as the same words: a 'muffin' is not the same as a 'puffin'. Do infants share this intuition? We present a new experimental method to test infants' spontaneous intuitions about phonological contrast.