Poster #116 - Children's Memory for Words and Speech Sounds Learned Via E-book Training

• In Event: 1-061 - Poster Session 02
  In Poster Session: PS 02 Section - Language, Communication

Integrative Statement

Infants and young children are able to learn and distinguish speech sounds efficiently via passive exposure. However, the hardest sounds to master are those that are not used in the learner’s native dialect (Escudero, 2005). We hypothesized that, with targeted speech-sound training, children can learn to distinguish speech sounds outside the range of their native dialect. This research program examines how children’s comprehension for words and speech sounds can be facilitated by repeated, passive listening to an electronic storybook (E-book) voiced in an unfamiliar dialect. The aim of this particular study is to identify whether such passive listening can benefit children’s discrimination of minimal pair words in an unfamiliar dialect.

In this study, preschool children (N = 36; M=41.2 months) from American English-speaking homes were presented with a pre-recorded story narrated in Australian English, a dialect that varies from American English in select vowels. The story was presented in an E-book format, which incorporates audio and visual (animated) stimuli presented on an electronic tablet (Figure 1). The E-book is ten “pages” long and contained four novel words and objects. Novel words included both minimal pair words (e.g., “det” and “dat”) and non-minimal pair words (e.g., “dat” and “sep”). Children were exposed to each of the novel words and objects five times during each E-book reading. Children were presented with two E-book readings in succession, followed by an immediate post-test to assess word learning and speech sound discrimination. At test, children are presented with twelve trials in which they observed a pair of novel objects from the E-book, and an experimenter then instructed children to select the referent for a target novel word (e.g., “Which one is det?”). Importantly, testing trials varied in the pairings of objects presented, such that some of the testing trials required children to discriminate among minimal pair words and some did not (Figure 2).

Results revealed that children’s memory for words differed across minimal- and non-minimal pair testing trials. There was a significant difference between the proportion of minimal pairs learned (M=.583) and non-minimal pairs learned (M=.694), t(35)=2.180, p=.036. In addition, children’s performance was above chance (.5) for non-minimal pair words, p<.001, while performance for minimal pair words was not significantly different from chance.

These results suggest that short, passive training exercises such as this can aid children in acquiring knowledge for words and speech sounds in non-native dialects, but only those which are less difficult to distinguish (i.e., non-minimal pair words). These findings have important theoretical value, as they suggest that children’s discrimination of certain speech sounds from non-native dialects can indeed be trained via passive learning. That is, these sounds are not entirely removed from speech perception within the first few years of life and can be accessed through listen to the sounds. This work also has implications for application to education, such as the development of young children’s language learning tools, including educational electronic apps and E-books.

Authors

• Catherine Bredemann, University of Wisconsin-Madison
  Presenting Author

• Haley Vlach, University of Wisconsin-Madison
  Non-Presenting Author

• Christopher Fennell, University of Ottawa
  Non-Presenting Author

• Clara C. Levelt, Leiden University
  Non-Presenting Author

• Liquan Liu, Western Sydney University
  Non-Presenting Author

• Paola Escudero, Western Sydney University
  Non-Presenting Author