There has always been a common-sense view that the number of languages that children learn, whether through natural exposure or educational intervention, has consequences for their development. The assumption was that these consequences were potentially damaging. Even now, after approximately 50 years of research on the topic, parents remain concerned about their children's development when it includes a bilingual experience. It is now clear that although parents were correct that speaking more than one language has consequences, the assumption about the nature of these consequences is not: the outcome of the experience is in fact the opposite of what many early researchers claimed and what many contemporary parents intuitively believe. In contrast to early warnings about negative consequences, bilingualism turns out to be an experience that benefits many aspects of children's development. Although there are documented delays in acquiring some formal aspects of each language, such as vocabulary (Bialystok Reference Bialystok, Luk, Peets and Yang2010), bilingualism has either no effect (intelligence) or positive effects (metalinguistic awareness, cognitive development) on development.

There are many reasons why the early research was consistent in its negative conclusions, but two of them are (a) the way in which the question was formulated and (b) the methodology used to conduct the research. Regarding the first, the early research focused primarily on intelligence as defined by standardized test scores. It is not at all clear what is measured by such tests, but we do know that IQ scores are strongly influenced by such factors as socioeconomic status (SES). Furthermore, IQ is a conglomerate measure that tells us little about specific aspects of intellectual functioning. In this sense, there is no compelling reason to expect that bilingualism per se, or at least bilingualism on its own, would influence IQ score outcomes. In the early studies, bilingual children actually scored lower than monolinguals on these tests, but it soon became apparent that the problem was in the second factor mentioned above, namely, methodology. In most studies, bilingual children were lower SES than monolinguals; when subsequent studies controlled for SES, IQ differences between groups disappeared. Similarly, bilingual children often have poorer verbal skills in the language of testing than monolingual children, a factor that was not considered in comparing performance on tests that typically have a high verbal component.

The field has progressed through the examination of different domains in search of bilingual effects on development. Following the early studies on measures of standardized intelligence, during the 1970s and 1980s interest turned to language acquisition and to metalinguistic development in particular. This was a sensible place to seek evidence of bilingual effects on development, and, indeed, the studies showed almost uniformly positive outcomes for bilinguals. Learning two languages in childhood changed the way in which children could think about language. Such metalinguistic insights are a central component of academic success in that they are in part what schools are designed to teach. Therefore, another line of research examined the effect of bilingualism on school achievement, in particular on literacy and mathematical abilities. Along with this, research began investigating the impact of new educational programs in which a language not necessarily used in the home was the medium of school instruction. Research in immersion education began in the late 1960s, but more focused studies on literacy and mathematical abilities were more prevalent in the 1990s.

An active focus of research since 2000 has been the effect of bilingualism on cognitive ability: specifically, the set of cognitive abilities known as the executive function. These are the processes responsible for attention, selection, inhibition, shifting and flexibility that are at the centre of all higher thought. Exciting new research is now providing strong evidence that early bilingualism has the power to set in place precocious development of these crucial skills.

Our review traces the main developments in this research by focusing on its evolution from the search for effects of bilingualism on intelligence, through interest in linguistic outcomes, to school achievement and, finally, to cognitive development. In some sense, the advancement of the field depended on the two points noted above, namely, getting the question right and designing an appropriate methodology. But there may have been a third factor responsible for the dire conclusions of the early research: there may have been a pre-existing bias to believe that bilingualism was harmful for children. Throughout each of the four thematic areas examined there is early evidence for either neutral or even positive outcomes of bilingualism, but negative claims typically continue for 20 years beyond those initial findings. Thus, for example, the result reported by Darcy (Reference Darcy1946) showing superior bilingual performance on non-verbal tasks, in conjunction with poorer performance on verbal tasks, is precisely the result found in the most current research, but the study was nonetheless used to support the view that bilingualism created a handicap for children.

The next stage in this research will be to engage in systematic studies of the consequences of bilingualism on structural and functional brain outcomes. This research is just beginning, and the next decade will take us to exciting new levels of understanding.

Our review is organized around the four major outcome variables examined in the literature we review.

1. A Intelligence

   1. A1: Verbal IQ test performance

   2. A2: Non-verbal IQ test performance

   3. A3: Creativity

2. B Metalinguistic awareness

3. C School achievement

   1. C1 Literacy

   2. C2 Mathematics

4. D Cognition

   1. D1 Theory of mind (i.e., the ability to attribute mental states to other people)

   2. D2 Executive functioning