Word extraction in adult-directed and infant-directed speech in African-American-English-learning infants

Although there is evidence that infant-directed speech (IDS) facilitates word extraction in Standard-American-English (SAE)-learning infants, this effect has not been tested in African-American-English (AAE)-learning infants. An important question that remains unaddressed is whether infants are better tuned to word boundary cues in their native dialect than a non-native dialect. Specifically, do infants extract words more efficiently from IDS in their native dialect than from IDS in a non-native dialect. This study compares the ability of eight-month-old AAE-learning infants to extract words from AAE and SAE IDS. The head-turn preference procedure was used to measure infants’ memory for nonsense words embedded in sentences in AAE versus SAE. Infants were familiarized for 2 min with six sentences containing nonsense words and were then presented with both familiar and unfamiliar nonsense words carrying matching prosodic features. The difference in looking times between the familiar and unfamiliar nonsense words was used as a measure of word extraction. Results so far demonstrate an advantage of IDS over adult-directed speech in SAE. IDS in AAE is less advantageous than IDS in SAE. Acoustic analysis shows that the prosodic features of target nonsense words in SAE may make them more salient to infants than target words in the AAE sample.