Language Acquisition
Infants face an enormous challenge in acquiring their first language. Language is extremely variable at every level of abstraction: the production of individual sounds and words will vary on a person-by-person and instance-by-instance basis; sentences can take on potentially infinite forms; and the goals and intentions of speakers frequently interact with and change the literal meaning of an utterance. Our research is interested in understanding how children solve this variability problem to acquire adult-like language behaviour. What kinds of information and biases might the child bring to the problem, and what information is available in the input itself to help solve it?
Size sound symbolism in mothers' speech to their infants
Point of Contact: Catherine Laing
Collaborators: Tamar Keren-Portnoy, Ghada Khattab, Shayne Sloggett
Description: Researchers have claimed that there is a relationship between the literal size of an object and the pitch with which expressions referring to that object are produced (i.e. smaller objects are realised linguistically with higher pitch). Moreover, it has been suggested that this feature of language may be used to facilitate word-learning in infants, with pitch acting as a meaningful cue for object/word pairs. This project questions the basis of this claim by examining the extent to which mothers routinely produce size-determined pitch when addressing their infants in naturalistic contexts. In other words, do mothers routinely modulate their pitch as a function of the size of the object they are referencing, or are previous findings the result of task-specific behaviours?
Phonological networks in development
Point of Contact: Catherine Laing
Description: Infants’ earliest words are structurally simple and phonologically similar. Data from a bilingual (English-Spanish) child’s early acquisition (Deuchar & Quay, 2000) shows that many of the child’s first words are phonologically identical: car, clock, casa ‘house’ and cat are produced as /ka/, and papa ‘daddy’, pájaro ‘bird’ and panda as /pa/. This suggests systematicity in the transition from babble to words, as infants draw on what they know to expand their early vocabulary using a small set of similar-sounding forms. This raises the question of how infants shift from these simple forms towards a more variable lexicon. To address this, this project draws on network analysis to identify changes in the child’s phonological system over the first three years of life: how similar are infants’ earliest words? And which factors influence the shift from similarity to variability? These questions are explored using corpus data from North American and French infants.