Relating Language Input to Language Processes Early in Development

• In Event: The Ins and Outs of Language Development: Integrating Multiple Measures from Inside and Outside the Laboratory

Abstract

Language development predicts later cognitive abilities. Thus, ensuring all children reach their full potential requires understanding what factors impact language development. For example, we know that children’s linguistic environments play an important role in their developing language abilities, but not many studies have looked at the relationship between language input and the underlying cognitive processes, such as word processing speed and referent selection, that support language development. Moreover, how changes in language input influence developing language processes at different time points is not clear, as longitudinal data relating children’s language environment to later linguistic processes are scarce.
We investigate the relationship between home language input at two different time points (6 and 18 months) and early language processes in a sample of UK infants. At each time point, we collected three days of LENA home recordings and extracted measures of adult word count, conversational turns, and child vocalisations, per hour. To measure language outcomes, we tested the same infants at 15 – 26 months-of-age, using the Early Language Processing (ELP) task. The ELP task uses a portable eye-tracker and established methods to measure multiple aspects of basic language processing.
ELP consists of two blocks of 34 and 29 trials separated by a 5-minute retention interval. On each trial, a pair of pictures is displayed on the screen for 2000 msec. Then a gaze contingent character appears in the middle and names the target. The pictures remain for a 3200 msec response period. There are four types of trials: speed of processing (highly familiar nouns that repeat during the task), comprehension (nouns, verbs and adjectives varying in difficulty), referent selection (one well-known and one novel object), retention (two previously-mapped novel objects).
Preliminary analyses show relationships between conversational turns (both adult and child initialised) at 6 months, trial type on ELP, and the age of the child at ELP (Figure 1). Children that did ELP before 20 months, show a benefit of more conversational turns on trials that contain novel words (referent selection and retention trials) whereas older children show a more general benefit of conversational turns on all but retention trials.
We also find relationships between conversational turns at 18 months and trial type on ELP (Figure 2) but, in this case, only turns initialised by an adult produce a statistically significant effect. In this case, children with more adult-initialised conversational turns do better at ELP, specifically in trials that contain highly familiar words such as speed of processing trials.
These results highlight the importance of conversational experience early in childhood. Also, they suggest that at different developmental stages, conversational turns might relate to different language processes. Analysis of the whole dataset (75 children) will look more in depth at the different ELP trials and will also explore how different language experience profiles (integrating children’s language input data at 6 months and at 18) relate to performance on ELP.

Authors

• Laia Fibla Reixachs, University of East Anglia
  Presenting Author

• John P Spencer, University of East Anglia
  Non-Presenting Author

• Larissa K Samuelson, University of East Anglia
  Non-Presenting Author