Home > Research > Publications & Outputs > Insights from studying statistical learning


Text available via DOI:

View graph of relations

Insights from studying statistical learning

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNChapter (peer-reviewed)peer-review

Publication date17/09/2020
Host publicationCurrent Perspectives on Child Language Acquisition: How children use their environment to learn
EditorsCaroline F. Rowland, Anna L. Theakston, Ben Ambridge, Katherine E. Twomey
Place of PublicationAmsterdam
PublisherJohn Benjamins Publishing Company
Number of pages25
ISBN (Electronic)9789027261007
ISBN (Print)9789027207074
<mark>Original language</mark>English

Publication series

NameTrends in Language Acquisition Research
PublisherJohn Benjamins
ISSN (Print)1569-0644


Acquiring language is notoriously complex, yet for the majority of children this feat is accomplished with remarkable ease. Usage-based accounts of language acquisition suggest that this success can be largely attributed to the wealth of experience with language that children accumulate over the course of language acquisition. One field of research that is heavily underpinned by this principle of experience is statistical learning, which posits that learners can perform powerful computations over the distribution of information in a given input, which can help them to discern precisely how that input is structured, and how it operates. A growing body of work brings this notion to bear in the field of language acquisition, due to a developing understanding of the richness of the statistical information contained in speech. In this chapter we discuss the role that statistical learning plays in language acquisition, emphasising the importance of both the distribution of information within language, and the situation in which language is being learnt. First, we address the types of statistical learning that apply to a range of language learning tasks, asking whether the statistical processes purported to support language learning are the same or distinct across different tasks in language acquisition. Second, we expand the perspective on what counts as environmental input, by determining how statistical learning operates over the situated learning environment, and not just sequences of sounds in utterances. Finally, we address the role of variability in children’s input, and examine how statistical learning can accommodate (and perhaps even exploit) this during language acquisition. © 2020 John Benjamins Publishing Company