Resolution of lexical ambiguity: evidence from an eye movement priming paradigm

Subjects' eye movements were monitored while they read 2-sentence passages of text. A target-word region was defined in the 2nd sentence of each passage. During the initial 35 ms of a target region eye fixation, an ambiguous word was presented as a prime. A target word subsequently replaced the prime during the fixation. Priming was measured by comparing fixation times on targets preceded by semantically related versus unrelated ambiguous primes. The type of prior context (consistent vs. inconsistent), type of ambiguous prime (biased vs. balanced), and strength of instantiated meaning (dominant vs. subordinate) could all affect priming. Priming effects were only found when the prior context was consistent with the dominant sense of a biased ambiguous prime. The results are discussed in terms of models of ambiguity resolution: The data seem most consistent with a reordered access model.