| Summary: | A study was conducted to evaluate orthographic neighborhood effects in parallel distributed processing models. Data were drawn from a set of stimuli that contained 2,073 words. Findings reveal that low frequency words with small neighborhoods had higher phonological, orthographic, and cross-entropy error scores than did low frequency words with large neighborhoods. It is also revealed that low frequency words without higher frequency neighbors had higher error scores than did low frequency words with higher frequency neighbors. It is concluded, as the result of an analysis of these models, that processing should be more rapid for low frequency words with higher frequency neighbors and for low frequency words with large neighborhoods.
|
|---|