Date created: | Last Updated:

Category: Project

Description: While prior studies have shown that monolingual infants can use transitional probabilities to segment words by 8 months of age (e.g., Saffran et al., 1996), less is known about how exposure to dialectal variation impacts this ability. The present study addresses this gap by examining how such variability influences word segmentation in 9-month-old Norwegian-learning infants. Specifically, three research questions were posed: (1) Do 9-month-old infants demonstrate statistical learning by segmenting words based on transitional probabilities? (2) Does exposure to two dialects influence segmentation compared to exposure to one dialect? (3) Among bidialectal infants, does the degree of dialect similarity affect segmentation performance? To explore these questions, the study implemented a conceptual replication of Saffran et al.’s (1996) artificial language paradigm. A total of 79 Norwegian-learning infants (43 monodialectal, 36 bidialectal) were familiarized with a continuous stream of synthesized speech containing four trisyllabic “words” defined by transitional probabilities. Infants’ ability to segment the stream was assessed using an eye-tracking task, with differences in looking time between familiar and novel test items serving as an index of segmentation.