NEIGHBORHOOD SIZE EFFECT IN NAMING - LEXICAL ACTIVATION OR SUBLEXICAL CORRESPONDENCES

Naming latencies are generally shorter for letter strings that are orthographically similar to many words than for those that have few lexical neighbors. Two experiments were conducted to explore the locus of the neighborhood size effect in a naming task. Experiment 1 showed that the neighborhood size effect was restricted to low-frequency words and was of comparable magnitude for words and for pseudowords. Experiment 2 examined performance as a function of the composition of the experimental lists. Target words were mixed with word or pseudoword fillers. The neighborhood size effect decreased in the pseudoword context relative to the word context. The data are compatible with the notion that the neighborhood size effect follows from the activation of specific word knowledge during print-to-sound computation.