Language Dominance Modulates Transposed-Letter N400 Priming Effects in Bilinguals

Abstract

Models of visual word recognition differ as to how print exposure modulates orthographic precision. In some models, precision is the optimal end state of a lexical representation; the associations between letters and positions are initially approximate and become more precise as readers gain exposure to the word. In others, flexible orthographic coding that allows for rapid access to semantics (i.e., 'good enough' orthographic processing) is the optimal end state. To adjudicate between these trajectories, we compared the size of transposed-letter ERP priming effects on two ERP components thought to reflect orthographic and lexico-semantic processing across languages in late English-Spanish bilinguals. Words that are represented precisely should be less susceptible to activation by transposed-letter primes (e.g., shpae-SHAPE) than words that are not, and should therefore yield smaller priming effects. Overall, targets elicited smaller N250s and N400s and faster responses when preceded by transposed-letter primes compared to substitution primes (e.g., shgue-SHAPE). The only effect that significantly differed between languages was N400 priming, which was larger in English, the dominant language. We suggest that these results favor models of learning to read according to which 'good enough' orthographic processing increases with print exposure.

The Neurophysiological Modality Effect in Native and Second Language Processing: An ERP Study

“…Namely, auditory information unfolds in a temporal, moment-to-moment order, whereas visual information becomes instantaneously available at the moment of presentation. Indeed, evidence based on the processing of transposed-letter nonwords (e.g., “cholocate” instead of “chocolate”) indicates that “good enough” orthographic representations of individual words are processed in much the same way as their correctly spelled versions (Lupker, Perea, & Davis, 2008; Meade, Grainger, & Holcomb, 2022). This suggests that visual words are processed holistically, at least to some extent, which, due to stimulus features, is not physically possible for auditory word processing.…”