Project on Children's Language Learning: Research Methods

Research Methods

Researchers in our laboratory employ several different techniques for examining children’s linguistic knowledge.

Preferential Looking

[Go back to the list of methods]

The Preferential Looking Procedure involves videotaping toddler's eye movements as they watch short movies on a large screen TV. The key idea behind this procedure is that children will look at images that match the speech that they hear. For example, if the screen has a picture of a dog on the right and a baby on the left, the infant will shift their gaze to the baby when they hear the question, "Where's the baby?" Children sit on a caregiver's lap, and the caregiver wears headphones so they do not influence the child's behavior. The movies typically last no more than 10 minutes. In our laboratory, we have been using this procedure to examine children’s early syntactic representations and how they use syntax to guide word learning.

Click here for a video example of a two-year-old participating in a study about verb learning. We present them with novel verbs to and compare what additional audio helps or hinders their learning of that verb. The corner video displays what the child sees on the television.

Click here for a video example of a two-month-old showing her preference to look at the face mouthing the sound she hears.

Truth-Value Judgment Task

[Go back to the list of methods]

Even though 3-4 year-olds have very sophisticated language abilities, they have difficulty introspecting about their language: they cannot verbalize what they know. The truth value judgment task is designed to determine how children interpret sentences without asking them directly. We tell children a story to set up a vivid context, and then ask them to judge whether a particular sentence is true or false with respect to what happened in the story. To keep children as young as 3 years old interested and attentive during our studies, we present the relevant contextual information using creative techniques like board games, animated videos or even iPad apps. We often have puppets utter the target sentences so children will feel comfortable saying some of the statements are wrong. In our lab, we have used this technique to examine children’s acquisition of the syntax and semantics of pronouns, quantifiers and attitude verbs.

Click to watch a video demonstrating the Truth Value Judgement Task.

"Every horse didn't jump over the fence." This is an example of an ambiguous sentence we can study using the Truth Value Judgement Task.

Meaning 1 (M1): None of the horses jumped over the fence.

Meaning 2 (M2): Some did jump and some didn't.

What do kids think? To find out, we tell stories where M1 is false and M2 is true. Watch the second video here.

This raised a new question: Can highlighting M2 in a different kind of sentence help kids access M2 in the ambiguous sentence? Our method was to tell the same sorts of stories, but our puppet says unambiguous "Not every --" sentences. Later, after a similar story, the puppet says the ambiguous "Every horse didn't -- " sentences. Watch the next video here.

Other child games

[Go back to the list of methods]

Often when investigating a new phenomenon, we discover that existing methods do not provide an appropriate measure of children's abilities. In these cases, we invent new games that we think give us a good chance of bringing out children's knowledge or the mechanisms used to acquire new knowledge.

Computational modeling

[Go back to the list of methods]

As we gather more information about children's language acquisition, it has become increasingly important to identify the mechanisms by which they construct their language. In recent years, we have been using computational modeling techniques in order to determine what information about linguistic structure might be present in the environment and how learners' initial knowledge makes that information usable.

Head-turn preference

[Go back to the list of methods]

This procedure involves measuring infants’ interest in different kinds of sounds. The infant sits face-forward on a caregiver's lap. The wall in front of the infant has a video monitor at eye-level. The walls on the sides of the infant have lights mounted at eye-level, and there are speakers beneath the lights. Sounds are played from the side speakers. The sounds start when the infant looks towards the blinking side light, and end when the infant looks away for more than two seconds. Thus, the infant essentially controls how long he or she hears the sounds. Differential preference for one type of sound over the other is used as evidence that infants can detect a difference between the types of sound. In our laboratory, we have been using this technique to examine children’s abilities to learn abstract syntactic and phonological generalizations in an artificial language.

  • Bergelson, E., & Idsardi, W. (2009). "Structural Biases in Phonology: Infant and Adult Evidence from Artificial Language Learning" BUCLD 33: Proceedings of the 33rd Annual Boston University Conference on Language Development, 85-96.

  • Takahashi, Eri (2009). "Statistical Learning of Hierarchical Phrase Structure in 18-Month-Old Infants". BUCLD 33: Proceedings of the 33rd annual Boston University Conference on Language Development. ed. Jane Chandlee, Michelle Franchini, Sandy Lord, and Gudrun-Marion Rheiner. Somerville, MA: Cascadilla Press.


[Go back to the list of methods]

During the Switch-Procedure, infants are seated on their caregiver's lap facing a large TV screen. Infants listen to different speech sounds, words or languages. At the same time, they look at a checkerboard or pictures of objects on screen. We measure their eye-movements as they look and listen. At the beginning of the experiment, infants are very interested in the speech sounds and so they direct their attention towards the television screen. Eventually, infants' attention decreases and they begin to look away. When infants reach this stage, we then change the speech. If infants notice the change and find it interesting, their attention is recaptured and they look back at the screen. In this way, we can see whether infants have detected the change in speech sounds, words or languages. In our lab, we have been using this technique to examine children’s acquisition of syntax and its role in word learning.