Article révisé par les pairs
Résumé : In Experiments 1-2, we replicated with two different Portuguese materials the consistency effect observed for French by Ziegler and Ferrand (1998). Words with rimes that can be spelled in two different ways (inconsistent) produced longer auditory lexical decision latencies and more errors than did consistent words. In Experiment 3, which used shadowing, no effect of orthographic consistency was found. This task difference could reflect the confinement of orthographic influences to either decisional or lexical processes. In Experiment 4, we tried to untangle these two interpretations by comparing two situations in which a shadowing response was made contingent upon either a lexical or a phonemic criterion. A significant effect of orthographic consistency was observed only in lexically contingent shadowing. We thus argue that lexical but not sublexical processes are affected by orthographic consistency.