In the last two decades there has been an explosion of research on bilingualism and its consequences for the mind and the brain (e.g., Kroll & Bialystok, Reference Kroll and Bialystok2013). One reason is that the use of two or more languages reveals interactions across cognitive and neural systems that are often obscured in monolingual speakers of a single language (e.g., Kroll, Dussias, Bogulski & Valdes Kroff, Reference Kroll, Dussias, Bogulski, Valdes-Kroff and Ross2012). From this perspective, the interest in bilingualism is about developing a platform to ask questions about the ways that cognitive and neural networks are engaged during language use, in different learning environments, and across the lifespan. Another reason is that an emerging body of research on the consequences of bilingualism suggests that language experience changes cognition and the brain (e.g., Abutalebi, Della Rosa, Green, Hernandez, Scifo, Keim, Cappa & Costa, Reference Abutalebi, Della Rosa, Green, Hernandez, Scifo, Keim, Cappa and Costa2012; Bialystok, Craik, Green, & Gollan, Reference Bialystok, Craik, Green and Gollan2009). Some of these changes have been claimed to produce cognitive advantages (see Bialystok et al., for a review of bilingual advantages and disadvantages).

In this keynote article, Valian (Reference Valian2014) provides a close analysis of one source of evidence that has been discussed and debated in this context. She focuses on studies across the life span that compare bilingual and monolingual performance on behavioral measures of executive function. The initial interpretation of the evidence for bilingual advantages in executive function was that the continual requirement to juggle the activation of the two languages conferred expertise that generalized beyond language to cognition more generally. But in Valian's review and comparison of executive function tasks across bilinguals of all ages, she argues that the evidence is inconsistent. Bilinguals sometimes outperform monolinguals and sometimes they don’t. She concludes that bilingualism contributes to cognitive functioning no differently from any other life experience.

There are many positive things to say about Valian's review and analysis. For those unfamiliar with the research on executive function, there is a useful review and appendix of tasks (although the work of Braver, Reference Braver2012, on reactive and proactive components of executive function is noticeably absent). There is also an informative review of some of the recent work on other types of expertise that have been hypothesized to confer related cognitive advantages.

The suggestion in this paper is that having a more detailed understanding of the components of executive function and their role in particular behavioral tasks will once and for all answer the question, of whether, and under what circumstances, there are cognitive benefits associated with multiple language use. In this brief commentary, I argue that this is a misguided approach. The evidence in the domain that is reviewed is indeed inconsistent. But the analysis itself is focused too narrowly. Without attention to how language itself is used and how the brain networks that support both language and cognition are affected by language use, we have an incomplete and potentially misleading picture of the problem. Valian (Reference Valian2014) notes that evidence has been reported for differences in bilingual and monolingual brains and suggests that there is analogous inconsistency in both the behavioral and neural data. But this misses the point that a much more complex analysis is required. That analysis will identify the cognitive and neural processes that are engaged by language use. A focus on inconsistency suggests that there is nothing particularly special about the mapping between bilingual language experience and its consequences. To the contrary, the recent evidence suggests that to investigate this relationship one needs a more sophisticated analysis that goes beyond the question of whether bilinguals differ from monolinguals on specific components of executive function to ask how bilingualism affects language processing and the cognitive and neural systems that support it.

A number of recent papers have attempted to provide a preliminary framework for investigating these issues (e.g., Baum & Titone, Reference Baum and Titone2014; Green & Abutalebi, Reference Green and Abutalebi2013). Most notably, Green and Abutalebi argue for what they call the adaptive control hypothesis, the idea that neural networks will be tuned according to the demands of language processing and language experience. Two bilinguals who are equally proficient may have acquired and may use the two languages in very different ways (and see Luk & Bialystok, Reference Luk and Bialystok2013). Some bilinguals code switch and others do not. Some bilinguals live in their native language environment and others in a second language environment. Some bilinguals use one spoken language and another signed language. The differences in these bilingual circumstances have been shown to affect the resulting cognitive consequences and provide a basis for generating specific hypotheses about their source (e.g., Emmorey, Luk, Pyers & Bialystok, Reference Emmorey, Luk, Pyers and Bialystok2008). Likewise, monolinguals will differ in all the sorts of life experiences described in Valian's (Reference Valian2014) analysis and there is frequently little attention paid to the variation in monolingual language experience that presumably contributes to some of the inconsistencies observed in the comparisons between bilinguals and monolinguals (e.g., Pakulak & Neville, Reference Pakulak and Neville2010).

I list below an additional, but partial, set of observations to illustrate the type of evidence that needs to be addressed to begin to allow the full and complex analysis that is required.