Although skilled bilinguals appear to be able to interact with others in whichever language is required, the evidence suggests that both languages are active when bilinguals read, hear, and speak one language alone.

By Judith F. Kroll, PhD

You are sitting at a café or at the airport when you overhear a conversation in English that suddenly switches to another language and then back to English. If you are a monolingual speaker of English, you may notice the mixture of languages without realizing that you have eavesdropped on an impressive feat of cognition. For many proficient bilinguals, code switching between two languages is a natural feature of language use (e.g., Myers-Scotton, 2002). Yet the same bilinguals rarely make the error of speaking the unintended language or speaking to a monolingual in a language that they don't understand. How is this cognitive control achieved?


Although skilled bilinguals appear to be able to interact with others in whichever language is required, the evidence suggests that both languages are active when bilinguals read (e.g., Dijkstra, 2005), hear (e.g., Marian & Spivey, 2003), and speak (e.g., Kroll, Bobb, & Wodniecka, 2006) one language alone. All of this happens even when the bilingual is highly proficient in both languages. It was once believed that these cross-language intrusions were present only during the early stages of learning a second language, when learners are necessarily dependent on transfer from their first language (e.g., Kroll & Stewart, 1994; MacWhinney, 1997). The new research shows that these interactions across languages persist and appear to be a feature of a system that is fundamentally permeable across language boundaries. The language not in use has been shown to influence bilingual performance at all levels, including the lexicon (e.g., Jared & Kroll, 2001), the grammar (e.g., Dussias, 2003), and the phonology (e.g., Sundara, Polka, & Baum, 2006). The presence of activity among both languages when only one language is required, in the absence of a serious disruption to performance, suggests that proficient bilinguals have acquired not only linguistic proficiency but also the cognitive skill that allows them juggle the two languages with ease. A particularly exciting development in the recent research is that the cross-language activity present for bilinguals appears to confer more general cognitive benefits in the realm of attention and executive function (e.g., Bialystok, Craik, Klein, & Viswanathan, 2004). A life experience of juggling two languages appears to create expertise in resolving cognitive conflict.

Evidence for cross-language activity

In Figure 1, we illustrate a Dutch-English bilingual deciding whether to call the pictured object of a bicycle a bikein English or afietsin Dutch. This seemingly simple decision may appear to be similar to decisions that monolinguals also experience. Should the object on which you sit in your living room be called a sofaor a couch? For monolingual speakers, choosing among close synonyms is a dilemma that occurs occasionally. For bilingual speakers, for whom most words in one language have a corresponding translation equivalent in the other language, the problem is rampant in the absence of a mechanism to easily switch off one of the two languages.

How can we demonstrate that the language not in use is available when bilinguals attempt to use one language alone? One research strategy is to identify properties of the bilingual's two languages that are ambiguous, in that they appear in both languages. If bilinguals are able to effectively switch off one of the two languages, then performance on language ambiguous words or structures should be no different than performance on unambiguous words and structures. The bilingual should be able to function as a monolingual. If the other language cannot be switched off, then performance for bilinguals should differ from performance from monolinguals on the same task. For example, in languages that share the same alphabet, it is possible to have words that are cognates in that they share the same or similar spelling and pronunciation and the same meaning. In Dutch and English, the word hotelis spelled identically and has the same meaning. However, it is also possible to have false friends or interlingual homographs which are words that also share the same or similar spelling and pronunciation but have different meanings in each language. In Dutch and English the word roomis an example of a false friend, meaning cream in Dutch.

Many recent studies have exploited cross-language ambiguities to determine how bilinguals are influenced by the presence of the ambiguity when attempting to understand these words in only one of their two languages. The results of these studies provide compelling support for the claim that bilinguals are affected by the language not in use. When cross-language form and meaning converge, bilingual performance is typically facilitated; when cross-language form and meaning conflict, bilingual performance is often hindered, in that it is slower and more likely to be error prone (e.g., Dijkstra, 2005). It might not seem surprising to find cross-language effects in the second or less dominant language since most bilinguals have one language that is more dominant. But recent studies show that the first language can be influenced by the second language at the level of the lexicon (e.g., Van Hell & Dijkstra, 2002) and also with respect to preferences for grammatical structures in sentence comprehension (e.g., Dussias & Sagarra, 2007). The bilingual is not only a mental juggler but also a language user who is not simply two monolinguals in the same mind (e.g., Grosjean, 1989; Malt & Sloman, 2003).

The role of context in constraining cross-language activity

One reaction to the observation of cross-language activity in word recognition is that real language experience rarely comes in the form of single words. Perhaps the evidence for parallel activity of the bilingual's two languages is a consequence of the decontextualized nature of experimental studies of word recognition. We normally speak and read sentences in the context of spoken discourse and written texts. The rich cues to the intended language should be available in context and might serve to accomplish the task of effectively switching off the other language.

A number of recent experiments have examined this question by asking whether the parallel activity of the two languages can be reduced or eliminated when the language ambiguous words that produce cross-language effects out of context, are placed in sentence context (e.g., Elston-Güttler, Gunter, & Kotz, 2005; Schwartz & Kroll, 2006; Van Hell, 1998).

In one set of studies, Schwartz and Kroll (2006) asked Spanish-English bilinguals to read sentences in their second language, English, in which particular words were either very likely to occur or not:
1. In the car my friend and I listened to songs on the radio and sang along.
2. My friend wanted to know if the radio we bought came with a warranty.
In the first sentence, the word radio is likely to occur in that position in the sentence. In the second sentence, many words could easily substitute for radio. In ordinary reading within the native language, readers are influenced by these constraints on sentence meaning. Critically, for bilinguals in Spanish and English, radio is a cognate, meaning that it shares spelling and pronunciation with its translation equivalent in the other language. Schwartz, Kroll, and Diaz (2007) showed that when bilinguals are asked to name a cognate like radio in isolation, they are faster relative to controls if the cognate shares both spelling and sound across the two languages. That result in itself suggests that even when reading in one language only, the other language is active. But what happens to the observed cross-language activity when bilinguals name these words in meaningful sentence context? When the sentence is highly constrained, as in the first example, the processing advantage for cognates disappears. However, in the sentence that is only generally constrained, an advantage for cognates remains, suggesting that knowing the language in which you are reading is not sufficient to switch off the influence of the unintended language.

Although research on cross-language activity in context is at an early stage, the results across a number of recent studies converge on the conclusion that context itself does not eliminate the activity of the language not in use. This finding is surprising in that we might expect bilinguals to exploit their knowledge of the language context to reduce the number of available interpretations. At the same time, it helps to explain why code switching may be prevalent and relatively cost free with respect to processing resources (e.g., Moreno, Federmeier, & Kutas, 2002).

Speaking two languages: A different problem?

Unlike reading or listening, speaking is a process that is necessarily initiated by a conceptual event such as a thought you want to express, a picture or scene to be described, or a word or sentence to be translated. Reading and listening engage bottom-up processes that are driven by the presence of text or speech. If the inability to switch off the unintended language is due to the data-driven nature of reading and listening, then it might be possible to show that bilinguals can indeed select the intended language when they plan spoken utterances in each of their two languages.

Quite surprisingly, the results on spoken production in two languages are consistent with the evidence on comprehension. Even the simplest production task, such as speaking the name of the bicycle shown in Figure 1, something that a three year old could do easily, appears to engage cross-language activity in the mind of highly proficient bilinguals (e.g., Costa, 2005).

Kroll, Dijkstra, Janssen, and Schriefers (in preparation) used a cued naming paradigm (see Figure 2) to examine cross-language activity during the production of words in each of the bilingual's two languages. Proficient Dutch-English bilinguals were presented with simple line drawings and were asked to name the picture aloud when cued by a high or low tone. In the mixed language conditions of the experiment, each tone cued one of the two languages. In the blocked language conditions, one tone cued one of the two languages and the other tone signaled a "no" response. In both mixed and blocked conditions, the tone was presented at a variable interval relative to the onset of the picture presentation.
The logic of this study was to use the comparison of naming performance under mixed vs. blocked language conditions as a way of determining whether both languages are normally active during speech planning. If both languages are active regardless of the requirement to use them, then performance under blocked conditions would be expected to resemble performance under mixed conditions. In contrast, if speech planning is fundamentally language selective, with activation of candidates only in the intended language, then mixed language conditions would be predicted to impose a significant processing cost for performance.

A comparison of naming performance under mixed and blocked conditions is shown in Figure 3. These data reveal a difference in the effect of language mixture for the first and second languages. For the second language, there was little consequence of language mixture, suggesting that even when only one language was required, both were active. In contrast, for the first language, there was a clear cost associated with requiring both languages to be active, suggesting that normally the first language can be produced without the influence of the second language. But perhaps the most striking feature of the data in Figure 3 is that when both languages were required to be used under the mixed conditions, the time to name pictures in the first language was longer than the time to name pictures in the second language. This reversal of the processing advantage normally seen for the dominant language suggests that the more dominant language was inhibited when the second language was also engaged.

Other studies have shown that words are available in both languages during production to the point where the phonology of the unintended language is on the tip of the bilingual's tongue (e.g., Costa, Caramazza, & Sebastian-Galles, 2000; Gollan & Acenas, 2004). In speech planning, the activated cross-language alternatives are likely to be translation equivalents and their spoken forms (e.g., the words bikeand fiets). In comparison, in comprehension, the activated alternatives are more likely to be lexical neighbors (e.g., the word bikein English may activate the word bijlin Dutch, which means axein English and therefore has little in common with bikeother than its word form). Thus, although the nature of what is active may differ in comprehension and production, in both cases there is clear support for the idea that both languages are available.

The consequences of cross-language competition

If both languages are active and compete for selection, the bilingual then needs to acquire a mechanism that provides a means to control this activity and the corresponding decision process so that random errors of language do not occur. At this point, there is a great deal of debate and discussion about what the control mechanisms might look like (e.g., Green, 1998) and how they might change with increasing proficiency in the L2 (e.g., Costa & Santesteban, 2004). Recent neuroimaging studies of bilingual performance converge with the behavioral results in suggesting that there is not evidence for a specific area of the brain associated with a language switch (e.g., Wang, Xue, Chen, Xue, & Dong, 2007). Other studies show that the requirement to select a single language to speak may engage a number of different brain areas associated with inhibitory control (e.g., Abutalebi & Green, 2007). A clear direction in the next stage of the research agenda will be to identify the specific components of inhibitory control that are associated with particular language processing tasks and with different degrees of success in second language learning.

We hypothesize that to become a successful speaker of two languages requires a fundamental reorganization of the entire language system, not simply the acquisition of isolated linguistic knowledge. That reorganization has consequences for language use in each of the bilingual's two languages and for cognition more generally. Bialystok et al. (2004) have shown that elderly bilinguals who have actively used their two languages their entire lives are offered a measure of protection again decline in executive function that appears to be specifically related to their bilingualism. It is tempting to speculate that these cognitive benefits are related to the competitive nature of language processing for bilinguals that we have reviewed in this article. The mental juggling that appears to be necessary to negotiate the use of two languages is a natural circumstance of bilingualism. Identifying the causal links that relate language processing to their cognitive consequences will be a critical next step in this research program. This area of research is relatively young but we believe that it holds great promise, not only for scientists interested in modeling competition across cognitive systems, but also for designing optimal learning environments for an increasingly diverse population of language users.

Abutalebi, J., & Green, D. (2007). Bilingual language production: The neurocognition of language representation and control.Journal of Neurolinguistics, 20, 242-275.

Bialystok, E., Craik, F. I. M., Klein, R., & Viswanathan, M. (2004). Bilingualism, aging, and cognitive control: Evidence from the Simon Task. Psychology and Aging, 19, 290-303.

Costa, A. (2005). Lexical access in bilingual production. In J. F. Kroll & A. M. B. De Groot (Eds.). Handbook of Bilingualism: Psycholinguistic Approaches (pp. 308-325). New York: Oxford University Press.

Costa, A., Caramazza, A., & Sebastian-Galles, N. (2000). The cognate facilitation effect: Implications for the model of lexical access. Journal of Experimental Psychology: Learning, Memory, and Cognition, 26, 1283-1296.

Costa, A., Hernandez, M., & Sebastián-Gallés, N. (2008). Bilingualism aids conflict resolution: Evidence from the ANT task.Cognition, 106, 59-86.

Costa, A., & Santesteban, M. (2004). Lexical access in bilingual speech production: Evidence from language switching in highly proficient bilinguals and L2 learners. Journal of Memory and Language, 50, 491-511.

Dijkstra, T. (2005). Bilingual word recognition and lexical access. In J. F. Kroll & A. M. B. De Groot (eds.), Handbook of bilingualism: Psycholinguistic approaches (pp. 179-201). New York: Oxford University Press.

Dussias, P.E. (2003). Syntactic ambiguity resolution in L2 learners: Some effects of bilinguality on L1 and L2 processing strategies. Studies in Second Language Acquisition, 25, 529-557.

Dussias, P. E., & Sagarra, N. (2007). The effect of exposure on syntactic parsing in Spanish-English bilinguals. Bilingualism, Language and Cognition, 10, 101-116.

Elston-Güttler, K. E., Gunter, T. C., & Kotz, S. A. (2005). Zooming into L2: Global language context and adjustment affect processing of interlingual homographs in sentences. Cognitive Brain Research, 25, 57-70.

Gollan, T. H., & Acenas, L. A. (2004). What is a TOT? Cognate and translation effects on tip-of-the-tongue states in Spanish-English and Tagalog-English bilinguals. Journal of Experimental Psychology: Learning, Memory, and Cognition, 30, 246-269.

Green, D. (1998). Mental control of the bilingual lexico-semantic system. Bilingualism: Language and Cognition 1, 67-81.

Grosjean, F. (1989). Neurolinguists, beware! The bilingual is not two monolinguals in one person. Brain and Language, 36, 3-15.

Jared, D., & Kroll, J. F. (2001). Do bilinguals activate phonological representations in one or both of their languages when naming words? Journal of Memory and Language, 44, 2-31.

Kroll, J. F., Bobb, S., & Wodniecka, Z. (2006). Language selectivity is the exception, not the rule: Arguments against a fixed locus of language selection in bilingual speech. Bilingualism: Language and Cognition, 9, 119-135.

Kroll, J. F., Dijkstra, A., Janssen, N., & Schriefers, H. (in preparation). Selecting the language in which to speak: Experiments on lexical access in bilingual production.

Kroll, J. F., & Stewart, E. (1994). Category interference in translation and picture naming: Evidence for asymmetric connections between bilingual memory representations. Journal of Memory and Language, 33, 149-174.

MacWhinney, B. (1997). Second Language Acquisition and the Competition Model. In A.M.B. De Groot & J. F. Kroll (Eds.),Tutorials in bilingualism: Psycholinguistic perspectives (pp. 113-142). Mahwah, NJ: Lawrence Erlbaum Publishers.

Malt, B. C., & Sloman, S. A. (2003). Linguistic diversity and object naming by non-native speakers of English. Bilingualism: Language and Cognition, 6, 47-67.

Marian, V., & Spivey, M. J. (2003). Competing activation in bilingual language processing: Within- and between-language competition. Bilingualism: Language and Cognition, 6, 97-115.

Moreno, E. M., Federmeier, K. D., & Kutas, M. (2002). Switching languages, switching palabras (words): an electrophysiological study of code switching. Brain and Language, 80, 188-207.

Myers-Scotton, C. (2002). Contact linguistics: Bilingual encounters and grammatical outcomes. Oxford, UK: Oxford University Press.

Schwartz, A. I., & Kroll, J. F. (2006). Bilingual lexical activation in sentence context. Journal of Memory and Language, 55, 197-212.

Schwartz, A. I., Kroll, J. F., & Diaz, M. (2007). Reading words in Spanish and English: Mapping orthography to phonology in two languages. Language and Cognitive Processes, 22, 106-129.

Sundara, M., Polka, L., & Baum, S. (2006). Production of coronal stops by simultaneous bilingual adults. Bilingualism: Language and Cognition, 9, 97-114.

Van Hell, J. G. (1998). Cross-language processing and bilingual memory organization. Unpublished dissertation, University of Amsterdam, Amsterdam, The Netherlands.

Van Hell, J. G., & Dijkstra, T. (2002). Foreign language knowledge can influence native language performance in exclusively native contexts. Psychonomic Bulletin & Review, 9, 780-789.

Wang, Y., Xue, G., Chen, C.S., Xue, F., Dong Q. (2007). Neural bases of asymmetric language switching in second-language learners: An ER-fMRI study. NeuroImage, 35, 862-870.

About the Author
Judith Kroll is Distinguished Professor of Psychology, Linguistics, and Women's Studies and Co-Director of the Center for Language Science at Pennsylvania State University. She completed her undergraduate work at New York University and her graduate work at Brandeis University. She previously held faculty positions at Mount Holyoke College, Rutgers University, and Swarthmore College. Together with Annette de Groot, she co-edited Tutorials in Bilingualism: Psycholinguistic Perspectives (1997, Erlbaum) and the Handbook of Bilingualism: Psycholinguistic Approaches (2005, Oxford). The research that she and her students pursue concerns the acquisition, comprehension, and production of words in two languages during second language learning and in proficient bilingual performance. Their work is supported by grants from the National Science Foundation and the National Institutes of Health and by a network of collaborators in The Netherlands, Spain, the UK, and China. Together with Suparna Rajaram and Randi Martin, she is one of the founding organizers of Women in Cognitive Science, an organization developed to promote the advancement of women in the cognitive sciences.