There has been much interest in when input variability aids learning of spoken language in both children and adults. We will focus on effects of variability in early language development, reviewing research to date on when phonetic variability facilitates language learning and when it does not. The structure is as follows: proposed mechanisms for facilitative effects of variability (e.g., Apfelbaum & McMurray, 2011); description of the types of variability that have been explored; review of evidence for variability effects. We will consider domains including phonological and phonotactic learning (e.g., Seidl, Onishi, & Cristia, 2014), word learning and word-form recognition (e.g., Rost & McMurray, 2009), and (briefly) accent (e.g., Creel, 2014). My hope is that reviewing the circumstances where variability aids learning during development will shed light on the mechanisms at play, to better understand how variability does (or does not) contribute to formation of robust, generalizable language categories.
We will explore evidence for and against the intriguing idea that infants may be characterized by greater learning flexibility than older learners, which may make them particularly well-suited for learning new language structure. We will review seminal work proposing age-related changes in learner characteristics (e.g., Newport’s 1990 “Less is More” hypothesis), alongside more recent evidence (e.g., Gerken, Goffman, & Quam, 2019). One domain we will emphasize in particular is developmental change in willingness to attend to non-native phonological dimensions during word learning (e.g., Singh et al., 2014). We will consider several possible mechanistic explanations for developmental changes in learning flexibility, including relative reliance on implicit vs. explicit learning (Quam, Wang, Maddox, Golisch, & Lotto, 2018) and increasing experience with native-language phonology. We will discuss implications for adult second-language learning and for developmental language disorders.