Measuring young children’s recognition of phonetically reduced word forms using story-guided looking

Infants readily recognize a variety of words when tested in the lab: e.g., when presented with a picture of a ball and a shoe, they look longer at the ball after hearing “Look at the ball!” (e.g., Fernald et al., 1998). This study develops a variant of this method to assess word recognition, embedding a standard fixation procedure (measuring children’s looking to one of two pictures given a sentence that aligns with only one of them) within the discourse context of a story. Inspired by work by Scott et al. (2012), who used a similar method to investigate theory-of-mind, we will present children with sequential sentences of a coherent story while two scenes are displayed visually, one matching the spoken sentence and one not. This new story-guided looking method provides a felicitous context to investigate children's response to properties of normal, conversational speech. In particular, we will examine young children’s understanding of reduced pronunciation variants. Reduction phenomena are pervasive in normal speech: Whole speech sounds can be dropped or added, or changed to incorporate features of nearby sounds (Lahey & Ernestus, 2014; Warner, 2019). Indeed, many instances of words in spontaneous speech are so different, or phonetically reduced, compared to the canonical form that they are phonetically uninterpretable in isolation (Bard & Anderson, 1983). This raises important questions about what young children, who are still learning the language they are trying to comprehend, understand about reduced pronunciation variants. In this study, we will test how well young children recognize reduced forms of common words and how the preceding context (presence or absence of a clear instance of the same word) influences recognition.