Phonological neighbourhood spread facilitates lexical decisions

Abstract

In visual word recognition, words with many phonological neighbours are processed more rapidly than are those with few neighbours. The research reported here tested whether the distribution of phonological neighbours across phoneme positions influences lexical decisions. The results indicate that participants responded more rapidly to words where all phoneme positions can be changed to form a neighbour than they did to those where only a limited number of phoneme positions can be changed to form a neighbour. It is argued that this distribution effect arises because of differences between the two groups of words in how they overlap with their neighbours.

Keywords:

• Phonological processing
• Phonological neighbourhood
• Visual lexical decision

Acknowledgments

I thank Chris Carbo for help with data collection. I thank James Adelman and an anonymous reviewer for their helpful comments on an earlier version of this manuscript.

Notes

¹ As a few of the words had high error rates, additional items analyses were performed after removing any item that had an error rate greater than 25%. This resulted in the removal of four additional P = 2 words and two additional P = 3 words. Based on this reduced set of stimuli, P = 2 words were still responded to more slowly and had a higher error rate (M = 671 and 5.9%) than P = 3 words (M = 633 and 3.1%). All F-tests that were reported as significant above remained significant with this reduced set.