This is an academia-to-English translation of a paper given to me by Erin Dowd - although to be honest, it's more a summary than anything else, since this paper is refreshingly well-written, at least compared to all the technical papers I've been reading lately - linguistics researchers appear to have better writing skills, on average, than engineering researchers. Taking notes here because I think the fact that people study these things is incredibly cool.
The Source of Enhanced Cognitive Control in Bilinguals: Evidence From Bimodal Bilinguals. Karen Emmorey, Gigi Luk, Jennie E. Pyers and Ellen Bialystok. Psychological Science 2008 19: 1201
If you speak two different languages (you're bilingual), older studies have shown you're probably better at making certain kinds of decisions - specifically, "executive control" problems where you have to pick one of multiple options. The guess is that bilingual folks need to choose between their two languages constantly, whenever they open their mouth, so they're used to that sort of mental switching.
But what if you know two different languages, but only one is spoken - so you don't have to choose what comes out when you open your mouth, because the other language uses your hands? That's what this study was about - there are folks like my friends Kevin Cole, Mackenzie "Maco" Morgan, and Steve Jacobs who both a spoken and signed language (ASL), so they're also bilingual. They can "simcomm," or sign-while-talking - two different modes of communication, so we call them "bimodal bilingual."
Contrast that someone who knows, say, Portuguese and English - obviously you can only speak one of those at a time, because you only have one mouth. Since bimodal bilinguals don't actually have to choose between their two languages while communicating - they can sign and speak simultaneously - do they also get the decision-enhancing benefits of being bilingual?
Nope. On the types of control tasks measured, everyone had about the same accuracy, but bimodal bilingual (signs-and-speaks) and monolingual (only-speaks-one-language) people took about the same amount of time to answer, whereas unimodal bilingual (knows-two-spoken-languages) people could do it faster. You've actually got to speak two different languages to reap these benefits.
Here end the actual notes on the paper. I'm personally curious what would happen if they studied (1) someone who knew two sign languages - say, Japanese Sign Language and ASL - you can't use two sign systems at the same time, so these folks would also be unimodal bilingual - and (2) someone who knew sign language and two different spoken languages (or perhaps if the simcomming folks were taught a second spoken language and then re-measured). These would arguably be difficult to find subjects for, and expensive to do (in the case of offering a language course to simcommers), but they make interesting thought experiments - my guess is that both of those kinds of subjects would get the "bilingual benefit," but we won't know unless someone actually tries it out.
Science! It's wonderful.