There is increasing evidence that in languages that have gender agreement, a congruent gender marking usually speeds up the processing of the following noun relative to an incongruent marking (or no marking). This effect is now well established in monolinguals, but little is known about how bilinguals react to gender agreement. In this paper, we ask whether bilinguals show the same effect and whether it depends on when they acquired and started using the gender-marking language on a regular basis.In what is fast becoming a classic, Corbett (1991) stated that gender is the most puzzling of the grammatical categories that interests nonlinguists as well as linguists, and that it becomes more fascinating the more it is investigated. Gender can be defined as follows: "A subclass within a grammatical class (as noun, pronoun, adjective, or verb) of a language that is partly arbitrary but also partly based on distinguishable characteristics (as shape, social rank, manner of existence, or sex) and that determines agreement with and selection of other words or grammatical forms" (Webster's Ninth New Collegiate Dictionary, 1991). Depending on the language, words (nouns usually) carry any number of genders; from two in such languages as Italian, Spanish, and French, all the way to six for Swahili. Of particular interest here is that other word classes in a language that has gender, such as adjectives, verbs, articles, pronouns, and so on, do not have a gender per se but can reflect, in their inflectional morphology, the gender of the words that do. Thus, depending on the language, a gender agreement marking can appear before or after a noun on a determiner, adjective, pronoun, and so on. In the case of French, for example, voiture is feminine, and in the phrase leur petite voiture (their small car), the adjective (petite) agrees with the noun and carries a feminine ending. In the phrase le garçon, the definite article is masculine since the noun is of that gender.It is now well established that a congruent gender marking will speed up the processing of the following noun. This has been shown in reading, for example, by Gurjanov, Lukatela, Lukatela, Savic, and Turvey (1985) in SerboCroatian and by Colé and Segui (1994) in French. In speech, Grosjean, Dommergues, Cornu, Guillelmon, and Besson (1994), working with French, showed with a gating task that participants needed less of a noun to identify it when it was preceded by a gender-congruent article (they were also more confident about the word they proposed), and with a lexical decision task, they showed that the participants were faster at deciding that the noun was a word. Recently, Jakubowicz and Faussart (1998), also working on spoken French and using a lexical decision task, replicated a strong gender effect.Researchers have also used a neutral or baseline condition in order to determine whether the effect is due to congruency (facilitation), incongruency (inhibition), or both. Schmidt (1986), for example, found a significant incongruency effect in German, but not a co...

In this paper we examine whether the recognition of a spoken noun is affected by the gender marking-masculine or feminine-that is carried by a preceding word. In the first of two experiments, the gating paradigm was used to study the access of French nouns that were preceded by an appropriate gender marking, carried by an article, or preceded by no gender marking. In the second experiment, subjects were asked to make a lexical decision on the same material. A very strong facilitatory effect was found in both cases. The origin of the gender-marking effect is discussed, as well as the level of processing involved-lexical or syntactic.