Panneton, R. 1 & Mills-Smith, L. 2
1 Virginia Tech
2 Roanoke College
Older infants learn object+label relations with maximal pairs, but not with minimal pairs (bin/din), because the latter is cognitively taxing. Will a female speaker delivering ostensive cues (ID speech; eye contact) enhance referential learning with minimal pairs? In Experiment 1, 20 14-month-olds eye-tracked a woman saying ?Hi Baby!?; after fixating her eyes, she turned and looked at an object on her left (or right), gazed at the infant, and labeled the object in IDS (8x). Next the woman repeated ?Hi Baby!?, looked at a second object (opposite side), gazed at the infant, and labeled the object in IDS (8x). During Test1, the woman looked at Object A and said Label B (switch trial); during Test 2, the woman looked at Object A and said Label A (control trial). Experiment 2 was identical except the woman looked away from the object being labeled. Only infants in Experiment 1 showed referential learning (increased attention to her mouth and to the object on switch trials). Thus, the intention of the speaker was a necessary aspect of successful referential learning although the same ostensive cues were present in both experiments. Moreover, enhanced, directed attention alleviated the cognitive load of minimal pair referential learning.