Length effects in word perception: We is better than I but worse than you or them.

Wheeler's finding that the words I and A do not have the same advantage as other words, despite their lexical status is replicated, and the strength and robustness of the word-length effect suggest that theories of theword advantage must include mechanisms that are length dependent.

Abstract

Under tachistoscopic viewing conditions letters are better reported if they are presented as part of a word than if they are presented in isolation. Several recent theories attribute this word advantage to the influence of lexical representations. The present study replicated Wheeler's finding that the words I and A do not have the same advantage as other words, despite their lexical status. This result holds even under conditions designed to influence subjects to process I and A as words. The poor performance on these single-letter words was shown to be a manifestation of a more general length effect: Recognition of briefly presented words improves with increasing length (up to three or four letters). The perceptual advantage for longer words was not found for closely matched strings of unrelated letters. The strength and robustness of the word-length effect suggest that theories of the word advantage must include mechanisms that are length dependent.