Phonological encoding in visual word recognition

Abstract

Three experiments use a tachistoscopic word recognition task to investigate how skilled readers convert visual input into a speech-related, or phonological code during reading. A specific model of phonological encoding (Spoehr & Smith, Journal of Experimental Psychology: Human Perception and Performance, 1975, 1, 21–34) is elaborated to provide the prediction that tachistoscopic report accuracy should be an inverse function of the length of the phonological code generated in processing the stimulus. Experiment I confirms the predicted phoneme-length effect while Experiment II shows that increases in syllable- and in phoneme-length both reduce report accuracy. Experiment III demonstrates that subjects' use of phonological encoding may change with varying reading conditions. The results of all three experiments are discussed in relation to specific models of reading.