Summer term 2017

04. May 2017:

Danielle Matthews (University of Sheffield)

How children break into language and become interesting talkers within 3 years

Within the first year of life, infants demonstrate the ability to communicate with others using gesture and vocalisations. Soon after, they break into conventional language and begin to use words. Having done so they begin to engage in proto-conversations that others naturally find interesting. The preschool years then form a period of forming and testing out expectations about how language works as a tool for facilitating the interactions they wish to have. By drawing on longitudinal studies of individual differences, corpus-based experiments and a randomized controlled trial, I will discuss how children make these qualitative leaps forward in language development and why a social gradient in language ability emerges around 18 months.

18. May 2017:

Eric Schulz (University College London)

Compositional inductive biases in human function learning

Function learning lies at the core of everyday cognition. From learning which stimulus will lead to reward all the way to how other people's intentions influence their actions, almost any task requires the construction of mental representations that map inputs to outputs. Since the space of such mappings is infinite, inductive biases are necessary to constrain plausible inferences. What is the nature of the human inductive biases over functions? How do people deal with complex functions that are not easily captured by standard learning algorithms? Insight into this question is provided by the observation that many complex functions encountered in the real world can be broken down into compositions of simpler form. We pursue this idea theoretically and experimentally, by first defining a hypothetical compositional grammar for intuitive functions and then investigating whether this grammar quantitatively predicts human learning, pattern completion, and memory and change detection performance. We end by speculating that compositionality is a necessary requirement for intelligent behaviour.

22. June 2017:

Christina Bergmann (Ecole Normale Supérieure de Paris)

Input variability in daily life: Does it matter who talks to infants?

In the last decades it has become clear that the speech input infants receive in daily life crucially shapes their individual paths into language: divergences in both quantity and quality of speech provided measurable individual differences on the levels of phonological processing and lexicon development. However, those studies typically focus on one main caregiver, whereas most infants grow up in much richer environments where multiple people interact with them and provide speech input. The presence of multiple speakers introduces variation on all linguistic levels, including those that are early milestones in languages acquisition, namely phonology and the lexicon. Predictions generated by concurrent theories (and backed by a few laboratory studies) cover all possible scenarios: It might be better to learn from fewer people to avoid confusion, more input speakers could be beneficial because infants are discovering abstract units in the speech signal earlier, or speaker differences might simply be irrelevant and infants ignore them. In short, speaker variability has different effects depending on the linguistic level considered and varying as a function of infant age (and thus presumably linguistic proficiency).