Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Babies, Brains, and Beats: Does the Way the Infant Brain Tracks Regular and Irregular Speech Impact Infant Word Learning?
Poster Session B, Wednesday, September 30, 4:30 - 6:30 pm, Wangari Maathai
Erica Flaten1, Janet Werker1; 1University of British Columbia
Even before they are able to produce any words themselves, 9-month-old infants can understand the meaning of many familiar nouns, and can also learn to associate unfamiliar syllables to distinct objects. Infants learn many of their first words from caregivers labeling objects in infant-directed (ID) speech, which is acoustically higher in pitch, slower in tempo, and more rhythmic (i.e., temporally regular) than adult-directed speech. Further, low-frequency brain activity (i.e., delta and theta) tracks the stress and syllable patterns in speech, a process called neural tracking. In general, for auditory rhythms such as speech or music, stronger tracking is often associated with better perception and/or memory, and stronger tracking is often found for rhythms that are more vs. less temporally regular. Infants show enhanced attention, learning, and neural tracking to ID than adult-directed speech, but whether rhythmic regularity, specifically, drives this remains unclear. We posit that rhythmic regularity engages infants’ attention and neural tracking, which thus enhances word learning. To test this, nine- to eleven-month-old infants (N = 47 with useable data so far, data collection is ongoing) from English-speaking homes participated in a word learning task with a familiarization phase and a test phase. To start, an actress appeared on the screen to introduce the task as a ‘naming game’ to the infants. Infants were then familiarized with two novel objects one at a time on the centre of the screen, where each was paired with a pseudoword (e.g., ‘Bap’ and ‘Det’), while eye-tracking, EEG, and video were recorded. Two additional words with familiar objects (‘Cat’ and ‘Dog’) were included to set the context of the task for the infants. Words were repeated by the actress (off screen) over an intonational phrase manipulated into patterns that were either rhythmically regular (fixed word inter-onset-intervals [IOIs] of 600 or 300 ms) or irregular (jittered IOIs ± 20-200 ms from regular patterns). At test, infants first saw the paired objects side by side without any audio to measure their baseline object preferences. Following this, they heard the taught pseudowords one at a time, repeated three times in a phrase (e.g., “Bap, look at the Bap! … Bap.”) while viewing both objects. Looking times to the correct vs. incorrect object after hearing each label in the test phrase were used to index learning. Preliminary analyses of infants’ looking times did not reveal above-change learning at the group level. At the same time, results for the neural tracking were surprising: infants’ neural tracking of speech in delta and theta bands did not significantly differ between regularity conditions, however, stronger tracking of the irregular phrases predicted better learning. These results suggest that although infants may not be showing behavioural evidence of learning word-object associations overall, there may be a separate role of neural tracking in this context, where infants who are better able to track the rhythmic content of the irregular rhythms benefit in learning. By combining behavioural and neurological measures, this study betters our understanding of how rhythmicity shapes early language acquisition.
Topic Areas: Language Development/Acquisition, Speech Perception