Abstract
Discrimination of a heard target word from phonologically similar and dissimilar items was investigated in seventeen 3.5- to 4.5-year olds and nine adults using a two-choice reaction time (RT) procedure. The target word (“little”) differed from the similar items (e.g., “lettle”) on a single acoustic-phonetic segment and from the dissimilar items (e.g., “cookie”) on all four acoustic-phonetic segments. Predictions based on the similarity neighborhood model (Luce, 1986) and the cohort model (Marslen-Wilson and Tyler, 1980) were (1) since word neighborhoods are less dense in children than in adults, children should show a smaller difference between their RTs to similar and dissimilar items than adults; (1) RTs for correct responses should be faster for dissimilar than for similar items; (3) children should false alarm more to similar than to dissimilar items; and (4) for the similar items, the later in the sequence of acoustic-phonetic segments the change occurs, the more easily confused the item should be with the target, generating higher false alarm rates. Results supported all predictions except the last, suggesting the cohort model may not provide an adequate description of young children's on line processing.
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call