This dissertation investigates the role of grammatical gender facilitation and inhibition in second language (L2) learners' spoken word recognition. Native speakers of languages that have grammatical gender are sensitive to gender marking when hearing and recognizing a word. Gender facilitation refers to when a given noun that is preceded by an adjective with the same gender marking (e.g., in Italian, bella casa "beautiful Feminine Adjective house Feminine Noun") is recognized more quickly than if the noun is preceded by an adjective that does not have a gender marking (e.g., grande casa "big Neutral Adjective houseFeminine Noun"). Gender inhibition refers to when a noun that is preceded by an adjective that has the wrong gender marking (e.g., *bello casa "beautiful Masculine Adjective house Feminine Noun") is recognized more slowly than if the noun is preceded by an adjective that does not have a gender marking (e.g., Bates, Devescovi, Hernandez, & Pizzamiglio, 1996; Dahan, Swingley, Tanenhaus, & Magnuson, 2001; Davidson, de la Fuente, Montrul, & Foote, 2011; Guillelmon & Grosjean, 2001). Effects of gender facilitation and gender inhibition among late second language (L2) learners are less strong than those found among native speakers (Davidson, et al., 2011; Foote, 2011) or nonexistent (Guillelmon & Grosjean, 2001). These earlier L2 studies, however, examined only learners whose first language (L1) was English, which does not have grammatical gender. In this study I compare the spoken word recognition of L2 learners with gender in the L1 to those without gender in the L1 in three different tasks (which differed in terms of automatic versus controlled processing) in order to investigate gender facilitation and inhibition in the L2 and how it is affected by the type of task and the transparency of the gender marking on the noun. A control group of 24 native speakers and 72 advanced learners of Italian divided in 3 experimental groups completed the three spoken word recognition tasks. The learners were placed into one of the 3 groups on the basis of gender in the L1: (a) Romance L1s that have gender systems similar to that of Italian (e.g., Spanish); (b) L1s that have gender systems different from that of Italian (e.g., Russian); and (c) L1s that do not have gender (e.g., English). The participants listened to adjective-noun combinations and responded by (a) repeating the noun, (b) by identifying the gender of the noun, or by (c) indicating the grammaticality of the adjective and the noun together. The response times and the accuracy of the participants' responses were recorded and analyzed with mixed measures ANOVAs and t tests. I predicted that the native speakers and all of the learners would show effects of inhibition in the more controlled explicit tasks and that the transparency of the gender of the noun would affect the speed with which they identified the nouns. I predicted that only the native speakers and the L1 Romance learners would show effects of facilitation and/or inhibition in the implicit, less controlled task. The results showed that learners of Italian who have Romance gender in the L1 are similar to native speakers in their processing of gender in the L2. Learners who have another (non-Romance) gender system in the L1 pattern more similarly to learners who do not have any gender in the L1, at least in controlled processing tasks. All participants were influenced by the transparency of the gender on the noun, but native speakers and L1 Romance learners were less affected than the other two learner groups. The findings are discussed in terms of the role that gender plays in the lexical access of L2 learners and how it differs from that of native speakers. The findings are also discussed in terms of implications for L2 acquisition of grammatical gender.
Read
