It feels like at some point in the recent past, the notion that being bilingual offers certain cognitive advantages (above and beyond allowing one to communicate in two languages) went fully mainstream. It almost seems obvious now: Well, duh, obviously exposure to a whole other language is going to change your brain in beneficial ways, specifically when it comes to focus and switching between tasks. But not so fast: A new metastudy in Psychological Science suggests that there’s some skewing going on when it comes to which sorts of studies on this subject are published — and, as a result, which ones trickle down to garner mainstream attention.
A team led by Angela de Bruin of Edinburgh University examined 104 papers on the bilingual advantage presented at academic conferences, which are often venues for researchers to get feedback on early results before refining their work and attempting to get published in peer-reviewed journals. The papers de Bruin’s team looked at were almost evenly split when it came to which ones had results supporting the notion of a bilingual advantage and which ones presented results challenging the notion (including “mixed findings” papers that leaned slightly in one direction or the other).
But then, when it came to getting these conference papers published in journals, something weird happened: 63 percent of the papers supporting a bilingual advantage achieved publication in a journal, while just 36 percent of those questioning the notion did. The team couldn’t find any obvious reasons that would be the case — it’s not like they had smaller sample sizes or anything like that.
What seems likely is that this is a case of publication bias — the tendency for scientific results that fail to find a link between two things (whatever the subject in question) to have a lot more trouble getting published than those that do find a link. In other words, all things being equal, many in the scientific community think that I’ll have a lot more luck publishing a paper showing that commonly used Substance A does cause cancer instead of one saying that there’s no evidence it does, even though both results are potentially important and meaningful (this is not a new notion for those who follow this stuff). So-called null results, a lot of science-watchdog types think, often get short-shrifted.
It’s easy to write all this off as nerdy nitpicking, but it matters a lot. After all, once results are published in big journals, they trickle down to mainstream publications (Science of Us included), and from there, they form the general societal sense of what is and isn’t true. A lot of people think the bilingual advantage exists because they’ve read about it in outlets they trust, and they may make certain decisions with regard to, say, raising their kids as a result. This metastudy doesn’t prove the bilingual advantage is a myth, of course, but it does suggest that the popular conception of it is built on a biased set of papers.
The researchers sum all this up nicely in their paper, as quoted in its press release: “All data, not just selected data that supports a particular theory, should be shared, and this is especially true when it comes to data regarding issues that have enormous societal relevance and implications, such as bilingualism.” As they say not only in English but in many other languages: Amen.
