Procedures we use to measures language learning in infants
We are able to measure language comprehension in infants using variety of methods. Because we can’t ask an infant what he or she hears when exposed to speech, we have to measure their level of comprehension indirectly. In our lab, we employ a technique called the Headturn Preference Procedure. In this procedure, the baby sits on a parent’s lap in a testing booth. On each side of the booth is a small light. From another room, a researcher makes one of the side lights flash, at which point speech proceeds to emanate from that particular wall. The researcher measures the amount of time the baby spends looking at light. The experimenter continues do flash the lights in conjunction with different types of speech, depending on the focus of the study. Typically, when infants recognize speech as being something they’ve already heard, they begin to look longer at the lights. In this way, we can obtain a measure of word recognition before a baby has uttered his or her first word. This technique has been used in laboratories to learn how much of what infants hear around them, they actually understand.
Below is a videotape of a infant participating in our study with her father. The videoclip is taken from a 15 minute session.
Procedures used to measure language learning in children
We are able to assess language in children (above 1 year) using standardized tests such as the Peabody Picture Vocabulary Test, the Expressive Vocabulary Test, and the MacArthur-Bates Communicative Development Inventory. All of these tests are normed.
In addition, to obtain a more naturalistic measure of language abilities, we typically hold a free play session where children are observed in conversation while playing with one of their parents. These sessions are audio-taped and transcribed for later analysis.
Privacy
We greatly appreciate the help that we receive from families in our research and make every effort to respect the privacy of families who participate. All data are treated anonymously and names and other identifying information is removed from the data files. Participants are assigned a code to protect their privacy. Data or names of participating families are never released to third parties. Data for an individual child is never analyzed in isolation and all data are collapsed into groups and analyzed as such.