Orthographic Neighbor Priming in Bilingual Sentence Reading: Effects of Individual Differences

Abstract

Previous research has demonstrated how form-similarity during visual language processing interferes with word retrieval and that the duration of this inhibition can be related to reading skills. The current study continued the research on orthographic neighbor priming in natural sentence reading. We recorded the eye movements of proficient Norwegian-English bilinguals during silent reading in their L2 (English). A target word was preceded by an orthographic neighbor or a control word and we also manipulated the distance between the prime and target (short vs. long) as well as the sentence structure (sentence break vs. no break). A key aim was to investigate the effects of individual differences, in this case L2 proficiency, working memory capacity and inhibitory control skills, on priming inhibition. Results from the experiment showed that neighbor priming effects are largely influenced by individual differences (L2 proficiency in particular) though these results were restricted to the short distance condition. Participants with lower proficiency and poorer inhibitory control skills demonstrated inhibition in the related condition. The data also showed stronger inhibition for individuals with better working memory capacity. In the sentence break condition, results showed that proficiency effects were cancelled across a sentence boundary, while working memory effects showed the opposite pattern from the no break condition when a full stop was included. In the long-distance condition, there was only a near significant effect of proficiency on relatedness, showing how more proficient bilinguals are more inhibited by form-similar words. Our findings provided further evidence for form-related priming inhibition during sentence reading and showed that these effects also apply when reading in a second language. The results also emphasized the importance of individual differences in studies on visual word recognition. The data is discussed in light of lexical competition models and the alternative episodic memory accounts.