| Popis: |
There is substantial evidence suggesting that infants as young as 8 months of age can use statistical information between syllables to segment speech (Karaman & Hay, 2018; Pelucchi et al., 2009a, 2009b; Saffran et al., 1996). However, the majority of research has only tested infants’ ability to segment words with perfect or near-perfect between syllable co-occurrence statistics (i.e., transitional probabilities ~ 1.0). Much less is known about how infants extract and represent syllable sequences with lower transitional probability (TP). In the current study, we seek to replicate and extend previous findings by having infants take part in a statistical learning task and examine their ability to extract both high-transitional probability (HTP; TP=1.0) and low-transitional probability (LTP; TP=0.3) words from a naturally produced Italian corpus. |
