A LOCUS OF THE WORD-LENGTH EFFECT ON WORD RECOGNITION

Abstract

Word length has a strong effect on visual word recognition. This study was conducted to examine the locus of the word-length effect by looking at the interacting pattern between word length and word frequency. If the interacting pattern is additive, the two variables are believed to affect separate stages. In contrast, the expanding pattern would indicate that at least some part of two effects arise in the same stage. A kind of preliminary experiment, Experiment 1, was performed to see whether the processing of word length and word frequency interact or not, using the naming task. There were more word length effects for the low-frequency word than the high-frequency word. This means the pattern of interaction was expanding. The main experiment, Experiment 2, was performed to see the interacting pattern between word length and word frequency, when the confounding variable, the recency, was controlled. The recency was controlled by repetition of all words. The interaction between length and frequency when recency was controlled was statistically significant. In addition, patterns of interaction between length and frequency did not change across the different recency conditions. Both Experiments 1 and 2 indicated that some part of the word-length effect and the word-frequency effect have the same locus. Because word frequency has the locus of a prelexical or lexical-access stage, some part of the word-length effect would occur in these stages.