Brief articleCognate effects in picture naming: Does cross-language activation survive a change of script?☆
Introduction
Past research suggests that for even highly proficient bilinguals the intention to speak in one language is not sufficient to restrict activation to that language alone (e.g., Colomé, 2001). Most of these studies have been performed in Dutch, English, Spanish, Catalan, and French with bilinguals whose two languages share the same Roman alphabet (e.g., Costa et al., 2000, Costa et al., 1999, Hermans et al., 1998, Kroll et al., in preparation). Only a few word recognition studies have considered the consequences of different-script bilingualism (e.g., Gollan et al., 1997, Jiang, 1999, Kim and Davis, 2003, Vaid and Frenck-Mestre, 2002).
The goal of the present experiment was to determine whether cross-language activation in spoken production is modulated for bilinguals whose languages use different scripts. Although the written form of a word is typically not present during spoken production, research on lexical processes suggests that all lexical codes are active to some degree regardless of the overt requirement for them to be used in a given task. In visual word recognition, phonology as well as orthography is active (e.g., Tan and Perfetti, 1999, Van Orden, 1987). In spoken word recognition, orthography is active even when the written word is not present in the task (e.g., Chéreau et al., 2007, Ziegler et al., 2003). Recent neuroimaging studies also show that although literacy is an acquired skill, the correspondences between letters and speech sounds are integrated in the same area of temporal cortex (Van Atteveldt, Formisano, Goebel, & Blomert, 2004). These findings converge with the results of bilingual word recognition studies showing that even among bilinguals who acquired their L2 after early childhood, there is interaction among the lexical codes associated with knowledge acquired in each of the two languages when only one language is required (e.g., Dijkstra, Grainger, & Van Heuven, 1999).
Other results show that orthography may be active in speech production. Damian and Bowers (2003) demonstrated that production is facilitated in a form-preparation paradigm when primes and targets are matched on both initial phonemes and graphemes, but not on either alone. Osborne, Rastle, and Burke (2004) replicated this finding in the picture-word Stroop task, showing that naming was facilitated only when the names of target pictures and distractor words were phonologically and orthographically similar. The implication is that all lexical codes are connected bi-directionally such that orthography modulates phonological processing during word production.
A question in the research on orthographic activation in speech production is whether the orthography itself needs to be present. Some studies suggest that the orthographic effect is found only when orthography is perceptually present in the task (Alario et al., 2007, Roelofs, 2006). However, it is important to note that these studies have been conducted in the first language (L1). Bilingual research suggests that L2 processing is slow and more permeable to the influence of other lexical information during speech planning (e.g., Kroll, Bobb, & Wodniekca, 2006) and that late bilinguals rely on transfer from L1 to L2 (e.g., Hernandez, Li, & MacWhinney, 2005). The extended time course of speech planning in the L2 and the activity of L1 make it more likely to reveal the effects of orthographic feedback if they are present.
If experience as a literate bilingual has consequences for the way in which the lexical codes associated with each language are activated and interact, then same and different-script bilinguals might be expect to produce different patterns of cross-language interaction. For same-script bilinguals, the presence of shared orthographic information would be expected to modulate the effects of cross-language phonology. For different-script bilinguals, cross-language interactions would be expected to depend on the activation of shared phonology only.
To test the hypothesis that script modulates cross-language activation in spoken production, we examined cognate facilitation in L2 picture naming. Past research has shown that bilinguals name pictures whose names are cognates in the two languages more quickly than pictures whose names are noncognates (e.g., Costa et al., 2000, Kroll et al., in preparation). Cognates are translations that are phonologically similar and potentially orthographically similar in same-script languages. The facilitation for naming cognate pictures suggests that lexical candidates in the unintended language are activated to the level of phonology during the planning of single word utterances. In the present study, we compared the performance of same-script bilinguals (Spanish–English) with that of different-script bilinguals (Japanese–English). If the phonology of both languages is activated even when L1 and L2 differ in script, then cognate facilitation should be observed for all bilinguals. However, to the extent that script serves to modulate phonological processing and/or as a cue to direct lexical access, then only Spanish–English but not Japanese–English bilinguals were expected to produce cognate facilitation.
Section snippets
Picture naming
Seventy-two black-and-white line drawings were sampled from Snodgrass and Vanderwart (1980) and Székely et al. (2003). Half of the pictures were cognates and the other half noncognates. Each cognate picture was matched to a noncognate picture on length of the picture’s name, frequency (Kucera & Francis, 1967), age of acquisition, name agreement, visual complexity (Székely et al., 2003), familiarity, imageability (Coltheart, 1981), variables shown to influence picture naming in past research
Latencies
The ANOVA revealed a significant main effect of cognate status (F1(1, 60) = 42.76, p < .001; F2(1, 66) = 6.74, p < .05). The main effects of cognate type and language group did not emerge (Fs < 1). The interaction between cognate status and cognate type approached significance by participants (F1(2, 120) = 3.06, p = .05), but not by items (F2 < 1). The interaction between cognate type and language group was reliable (F1(2, 120) = 7.02, p;< .01; F2(2, 66) = 4.81, p < .05). There was no interaction between cognate status
General discussion
The goal of this experiment was to determine whether cross-language activation is modulated by script differences when the written lexical form is not present. The results replicated Costa et al., 2000, Kroll et al., in preparation and extended the finding of cognate facilitation to different-script bilinguals. The monolingual data did not reveal any of these effects. Our findings suggest that even when the bilingual’s two languages do not share script, there is activation of the phonology of
Acknowledgements
This research was supported by NSF Grant BCS-0418071 to JFK and by NSF Dissertation Grant BCS-0518814 to NH and JFK. We thank Teresa Bajo and Hiroko Tajika for the use of their laboratories, Judith Pirela, Matilda Navarro, James Burns, and Lindsey Neblock for research assistance, and Albert Costa, Ton Dijkstra, Jared Linck, Pedro Macizo, Maya Misra, Tyler Phelps, and Janet Van Hell, for helpful comments on an earlier version of the manuscript.
References (30)
- et al.
The role of orthography in speech production revisited
Cognition
(2007) - et al.
Reading spoken words: Orthographic effects in auditory priming
Cognition
(2007) Lexical activation in bilinguals’ speech production: Language-specific or language-independent?
Journal of Memory and Language
(2001)- et al.
Lexical selection in bilinguals: Do words in the bilingual’s two lexicons compete for selection?
Journal of Memory and Language
(1999) - et al.
Recognition of cognates and interlingual homographs: The neglected role of phonology
Journal of Memory and Language
(1999) - et al.
Effects of orthography on speech production in a form-preparation paradigm
Journal of Memory and Language
(2003) - et al.
The emergence of competing modules in bilingualism
Trends in Cognitive Sciences
(2005) - et al.
Task effects in masked cross-script translation and phonological priming
Journal of Memory and Language
(2003) - et al.
Do orthographic cues aid language recognition? A laterality study with French-English bilinguals
Brain and Language
(2002) - et al.
Integration of letters and speech sounds in the human brain
Neuron
(2004)
Neighborhood effects in auditory word recognition: Phonological competition and orthographic facilitation
Journal of Memory and Language
Predictors of picture naming speed
Behavior Research Methods, Instruments, & Computers
The MRC Psycholinguistic Database
Quarterly Journal of Experimental Psychology A
The cognate facilitation effect: Implications for models of lexical access
Journal of Experimental Psychology: Learning, Memory, and Cognition
The architecture of the bilingual Word recognition system: From identification to decision
Bilingualism: Language and Cognition
- ☆
This manuscript was accepted under the editorship of Jacques Mehler.