Research interests

Words are typically not uttered in isolation but rather embedded in a stream of speech. This influences language acquisition and processing in at least two major ways. The first of these is the fact that the young language learner must be able to extract recurring sequences of sounds (i.e., words) from a fluent speech stream often without prior knowledge of what these words are and, thereafter, assign meaning to these sequences. Our research examines the factors that aid this process, focussing both on extrinsic factors - such as the role of the input to the child - as well as intrinsic factors - such as the child's own interest in learning - in driving language learning. This includes analysis of the acoustic characteristics of the input, e.g., the extent to which the input is appropriately child-directed,  the influence of the infants’ own interest in language on learning, the influence of co-occurring manual gestures on learning and the level of individual variation in the quality and quantity of the input to infants. In sum, this work focuses on identifying the linguistic and social characteristics of our interactions with children that maximally benefit language learning.

The fact that words are typically uttered in fluent speech also impacts language processing in another way, namely, that these surrounding words provide a context in which incoming words are processed. Our work examines the role of this surrounding context on language processing in language learners (toddlers, children acquiring literacy, bilingual children and adult second language learners) and the extent to which this surrounding context speeds day-to-day language processing. In particular, we examine, for instance, whether language learners retrieve the phonological and semantic properties of words online during word recognition, whether this online retrieval facilitates word recognition and whether learners can use this retrieved information to predict upcoming language and the extent to which the characteristics of the young learners’ lexicon influences such retrieval across development. 

Research methods

Most of our research involves one of two methods, i.e., the eye-tracking methodology and the event-related potential methodology. For eye-tracking studies, we typically present infants with images of objects and words referring to these objects and follow their eye-movements across the objects presented to them. We use the amount of time they fixate an object as an index of their association of a sound with an object. We also use the Event-related potential technique, a safe and non-invasive procedure, where we measure the normally occurring activity in the baby’s brain associated with their recognition of a picture or a sound presented to them.