You are viewing the SNL 2017 Archive Website. For the latest information, see the Current Website.

 
Poster C22, Thursday, November 9, 10:00 – 11:15 am, Harborview and Loch Raven Ballrooms

Insight into spoken word processing in young children using eye movements

Elizabeth Simmons1,2, Rhea Paul3, Rachel Theodore1,2, Monica Li1,2, James Magnuson1,2;1University of Connecticut, 2CT Institute for Brain and Cognitive Sciences, 3Sacred Heart University

Models of spoken word recognition (SWR) in adults highlight the importance of temporal order (Marslen-Wilson, 1987; McClelland & Elman, 1986) for recognizing words. Words that overlap completely at onset (cohorts) or mismatch only at onset (rhymes) compete for recognition as a function of phonetic similarity over time, with stronger, earlier competition from cohorts than rhymes (Alloppenna, Magnuson, & Tanenhaus, 1998). Theories make conflicting predictions about the relative time course of cohort and rhyme competition in very young children (pre-readers), but to date, they have not been assessed empirically. Some scientists have theorized that early on, children have underspecified representations, and process words in a holistic fashion, paying minimal attention to the sequence of phonemic information and that a gradual refinement of these representations takes place through age seven (Walley, 2003; Treiman & Baron, 1981). On this view, degree of competition would depend on degree of overlap, not temporal position of overlap between words. Another view is that phonemic-grained sequences are not accessible prior to explicit reading instruction (Liberman, Shankweiler & Liberman, 1989). Both theories predict that very young children who have not yet learned to read will not demonstrate adult-like patterns of cohort and rhyme competition. An alternative possibility is that patterns of cohort and rhyme competition observed in adults emerge as a natural consequence of processing words over time (Magnuson, Tanenhaus, Aslin, & Dahan, 2003). This study attempts to distinguish among these possibilities by measuring for the first time the fine-grained time course of SWR in a group of preschool children who are not yet reading. Twenty-three monolingual, typically developing preschool children (3-5 years) completed an eyetracking task using a developmentally-appropriate extension of the visual world paradigm (Tanenhaus et al., 1995). Participants heard a simple auditory instruction ("find the coat") and were asked to use a mouse to click on the named picture. The target object (e.g., coat) was paired with either a cohort competitor (e.g., comb), rhyme competitor (e.g., boat), or a phonologically unrelated object (e.g., bird). Participants’ eye movements were recorded as fixations to the target or the competitor. Our results revealed that the time course of activation and competition among words was similar to that of adults (Allopenna et al., 1998), even among the youngest participants. While children exhibited increased latency compared to adults, competition effects were similar: cohorts competed early and strongly, while rhymes competed more weakly with a later peak, but were still significantly more likely to be fixated than unrelated words. These findings support the notion that even young children, who have yet to be exposed to formal reading instruction, have access to differentiated lexical representations and utilize subtle, temporal differences in the signal to access spoken words.

Topic Area: Language Development

Back to Poster Schedule