On the effects of second language immersion on first language production Cristina Baus a, b, ⁎, Albert Costa a, c , Manuel Carreiras b, d, e a Center of Brain and Cognition (CBC), Universitat Pompeu Fabra, Barcelona, Spain b Basque Center on Cognition, Brain and Language (BCBL), Donostia, Spain c Institució Catalana de Recerca i Estudis Avançats (ICREA), Barcelona, Spain d Ikerbasque, Basque Foundation for Science, Spain e Departamento de Filología Vasca, Universidad del País Vasco, Spain abstract article info Article history: Received 8 June 2012 Received in revised form 16 January 2013 Accepted 18 January 2013 Available online 24 February 2013 PsycINFO classification: 2340 Keywords: Bilingualism Speech production L2 immersion Cognates Lexical frequency The present study was aimed at investigating how the effects of a second language (L2) immersion on first language (L1) production are modulated by lexical frequency and the cognate status of words. By means of a picture naming and a semantic fluency task, we longitudinally explored the changes in L1 performance of a group of immersed learners. Our results revealed slower naming latencies at the end than at the beginning of the immersion period but only for those pictures whose corresponding names were low frequency and non-cognates. Moreover, the semantic fluency task revealed a decrease in the percentage of non-cognate words produced at the end of the immersion period. The observed decline in the accessibility to L1 words after a short L2 immersion period is explained in terms of the general bilingual lexical access disadvantage and studies of L1 attrition in speech production. Specifically, the present findings are more in line with theoretical accounts that consider lexical difficulties to be the result of changes in the frequency of use of the L1 during a short L2 immersion period. © 2013 Elsevier B.V. All rights reserved. 1. Introduction Despite the practical benefits of knowing more than one language, bilingualism seems to also entail certain linguistic costs. As compared to monolinguals, bilinguals suffer a disadvantage in word retrieval during speech production even in their first language (L1). For instance, when speaking in L1, bilinguals are slower and produce fewer words than monolinguals (Gollan, Montoya, Fennema-Notestine, & Morris, 2005; Gollan, Montoya, Cera, & Sandoval, 2008; Gollan et al., 2011; Ivanova & Costa, 2008; Portocarrero, Burright, & Donovick, 2007; Rosselli et al., 2000; Sadat, Martin, Alario, & Costa, 2012; Sandoval, Gollan, Ferreira, & Salmon, 2010; but see Ransdell & Fischer, 1987). Also, long and in- tense exposure to a foreign language accompanied by a reduction in the use of the L1 may lead to the phenomenon of language attrition (De Bot, 1999; Keijzer, 2007; Schmid, 2007; Seliger & Vago, 1991). The effects of changes in the frequency with which the two lan- guages are used seem to appear relatively quickly. As demonstrated both in a laboratory context (Jared & Kroll, 2001; Levy, McVeigh, Marful, & Anderson, 2007) and more natural L2 learning settings (i.e., L2 immersion context; Linck, Kroll, & Sunderman, 2009), difficul- ties in L1 lexical retrieval appear shortly after intensive L2 practice. For example, Linck et al. (2009) have shown that English native speakers learning Spanish in an immersion context for one semester produced fewer exemplars in a fluency task in their L1 (English) than classroom learners without immersion experience. In the present study, we aim at further characterizing this effect of language immersion on the availability of L1 lexical representations by complementing Linck et al.'s findings in several ways. Firstly, we aim to explore immersion effects on the L1 from a different experi- mental approach: by assessing across the time of immersion the longitudinal changes in naming performance of the same group. This methodology (together with the transversal methodology used in Linck's study) will help us to provide a more complete picture of how immersion affects accessibility to L1 words. Secondly, we aim to explore the effect of lexical properties that are known to impact the availability of lexical representations in L1: lexical frequency and cognate status. Assessing how these variables affect the availability of L1 lexical representations after a L2 immersion period is important because it will not only help to characterize this phenome- non better but may also help us to understand its origins. In particular, our study may shed some light on whether the effects of L2 immersion in L1 (and more in general the bilingual disadvantage) are due to a reduction of L1 use or due to the potential interference produced by the L2. At present, there is little experimental evidence about how these two variables affect L1 processing in cases of L2 immersion. In the following, we describe the main findings about frequency and cognate effects in bilingual contexts and theoretical accounts of Acta Psychologica 142 (2013) 402–409 ⁎ Corresponding author at: Department de Tecnología, Universitat PompeuFabra, C/Tànger, 122-140, 08018, Barcelona Spain. Tel.: +34 93 542 13 82; fax: +34 93 542 25 17. E-mail address: baus.cristina@gmail.com (C. Baus). 0001-6918/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.actpsy.2013.01.010 Contents lists available at SciVerse ScienceDirect Acta Psychologica journal homepage: www.elsevier.com/ locate/actpsy