The time course of graphic, phonological, and semantic activation in Chinese character identification.

TLDR

Lexical form–form relations are often more reliable than form–meaning relations, so phonological forms activated by graphic forms become constituents of word identification, implying phonological access can precede meaning access in single‑word reading. The authors tested Chinese word reading in two primed‑naming experiments that varied prime type and prime‑target SOA, and proposed an interactive constituency model to explain the observed activation patterns. The experiments revealed a temporal sequence of facilitation—graphic first, then phonological, then semantic—where precise meanings yielded faster semantic priming, graphic prime facilitation at 43 ms turned to inhibition at longer SOAs, and the onset of inhibition matched phonological facilitation, indicating a single identification moment.

Abstract

In reading, lexical form-form relations may be more reliable than form-meaning relations. Accordingly, phonological forms (activated by graphic forms) become actual constituents, rather than addenda, of word identification. These considerations suggest that access to phonological forms can precede meaning access in single-word reading in many circumstances. The time course of form and meaning activation during Chinese word reading was tested in 2 primed-naming experiments varying prime type and prime-target stimulus onset asynehrony (SOA). The results showed a sequence of facilitation over SOA: (a) graphic, (b) phonological, (e) semantic. Words with precise meanings produced more rapid semantic priming than words with vague meanings. Graphic prime facilitation at a 43-ms SOA gave way to inhibition at longer SOAs. The onset of graphic inhibition coincided with the onset of phonological facilitation, suggesting a single identification moment. The authors describe an interactive constituency model that accounts for the pattern of data.