Learning first words

The researchers in this project are studying how “hyperarticulation”, that is, extreme or exaggerated articulation, impacts three aspects of language development, all of which are needed so that an infant can learn new words. 

The three aspects are word recognition, i.e. recognizing and understanding the word; word segmentation, i.e. being able to know where one word ends and another begins in normal speech without pauses; and word learning, i.e. the specific ability to associate a word with an object or concept.

Earlier research has shown that the specific way of speaking that people use with babies – “baby directed talk” – has a positive impact on these three aspects of early language learning. It has also been found that the amount of baby directed talk in a child’s day-to-day surroundings correlates with later language abilities. 

The reason the researchers are focusing on hyperarticulation, which is one of several distinguishing features of baby directed talk, is that its function in conversations between adults is to increase the chances of successful communication. When we speak we constantly adapt our speech style, and when we have reason to suspect that our message is not getting through – due to background noise, for instance, we tend to exaggerate our articulation to make it easier for the listener to understand what we are saying. The researchers think that these intuitive adaptations highlight information and therefore help the baby to listen for what is relevant to understand, learn and recognize words. 

The researchers will be using eye-tracking to record children’s reactions when they see and hear different images and speech sequences to find out whether hyperarticulation influences children’s ability to understand words, find word boundaries and form associations between words and objects. They will be examining the impact of the degree of hyperarticulation, variation in hyperarticulation, and how relevant hyperarticulation is in the context.

This project will help us to learn about how infants learn to understand and use language. This knowledge will also be useful in other research fields, since the principles underlying the learning of new words can be extrapolated to form general principles applying to the processing of information. This means the project can also contribute to fields as disparate as speech and communication technology, preschool pedagogy and speech therapy.

Project:
Learning first words: The effect of hyperarticulation on infant word-recognition, word-segmentation and word-learning

Principal investigator:
Lisa Gustavsson

Co-investigators:
Ellen Marklund
Iris-Corinna Schwarz

Institution:
Stockholm University

Grant:
SEK 4 million