Friday, March 31, 2006

Bilingualism = Multiple Personality Disorder?

OK, this (via Mind Hacks) is just cool, even if I don't know quite how seriously to take it. Apparently there's been a fair amount of research over the last few years on changes in attitudes, values, and behaviors that occur when bilinguals (or multilinguals) switch between languages. For example, in one study (p. 2):
Hong Kong bilingual-Chinese managers who responded to a values questionnaire in English displayed means closer to a group of American managers in the US than did the bilingual-Chinese managers who responded to the same questionnaire in Chinese (Ralston et al., 1995).
The explanation, which the authors of the linked study (two of whom I know pretty well personally, and still had never heard of this stuff) attribute to "Cultural Frame Switching" or "cultural accomodation," is that the language primes the culture that goes with it, and the cultures values and atitudes are thus primed as well. You know, I can almost buy that.

But for social psychologists, the mere priming of values and attitudes is not sexy enough. They need something bigger; they need to show that switching between languages causes personality changes. Of course, this requires showing something equally sexy, namely that differences in personality exist between Spanish and English speakers in the first place. So the paper is, like, doubly sexy ("sexy" is, of course, a technical term in social psychology, and doesn't refer to anything related to actual sex... I hope).

In the first study, they administered the Big Five Inventory, which measures the Big Five personality dimensions, to 168, 451 (yes, that's a lot) English-speaking participants in the United States, and 1031 Spanish-speaking participants living in Mexico. They found small, but statistically different differences on all 5 dimensions (with samples that large, it's no wonder), such that the English-speakers scored higher on the Extraversion, Agreeableness, Conscientiousness, and Openness dimensions, while the Spanish-speakers scored higher on the Neuroticism dimension.

Using this finding, they then predicted that if bilinguals take the BFI in both languages, they will score higher on Extraversion, Agreeableness, Conscientiousness, and Openness when they take it in English, and higher on Neuroticism when they take it in Spanish. In three studies, with a combined sample of 249 English-Spanish bilinguals in Mexico and the United States, they found differences in the predicted directions on four of the five dimensions, with the difference on the Openness dimension being in the opposite direction, though not statistically significant. The difference between the scores on the Spanish and English versions on the Neuroticism dimension, while in the predicted direction, was also not statistically significant.

Once again, the authors interpret these results in terms of Cultural Frame Switching. They also make it clear that, while they found small changes in personality within individuals, the correlation between scores on the English and Spanish tests, across all participants, was 0.8. This implies that the differences between the scores for a single individual on the two versions of the test were relative. If one person takes the test in English and scores higher on the Extraversion dimension than another person did on the English version, it's likely that the first person will also score higher on the Extraversion dimension on the Spanish version than the second will on the Spanish version. Thus, switching languages isn't really altering your personality all that much. It's just tweaking the levels a little. And switching between Spanish and English doesn't seem to change your level of openness at all. I guess openness isn't as prone to priming effects as, say, extroversion. So, if you're an overly close-minded Spanish speaker, I'm afraid that learning English and speaking it exclusively probably won't help. But if you're an introverted Spanish speaker with a messy bedroom (Conscientiousness), and you want to have more fun at parties and finally clean that room, then by all means, learn English.

13 comments:

Jeff G said...

Actually, I kind of suspected this, although I would imagine that the subjects have to be REALLY bilingual rather than just having studied another language at some point. The reasoning which lead me to suspect this before I even read this post came from my belief/suspicion that morals and values are simply institutions and institutions are simply part of our language. Thus different languages would entail the possibility, perhaps even probability, of a difference in values and moral to a certain degree.

Chris said...

Yeah, I have no doubt that languages can prime values and attitudes. Of course, you'd have to be more than bilingual. You'd also have to be very knowledgeable about the two different cultures. But personality? OK, that's a little more difficult to swallow. I don't doubt their data, and the samples seem to be OK. What I wonder is how long such an effect (and the effect sizes were all pretty small) would last, and what the practical impliciations might be.

Heo said...

Hey, I know two languages and that test suggests that I have no personality. I've been gyped!

Chris said...

Heo, that's because Old English speakers in... um... Old England are personalityless.

Anonymous said...

How do you correct for the fact that the questions are necessarily different in the different languages?

It'd be interesting if you could do a language-neutral personality test, with images or something, and have people take it while thinking in or after speaking in each language.

Chris said...

Anon, that's an excellent question, and I don't say that just because it was the first thing I thought when I read the paper's abstract (OK, maybe that's part of why I say it). That's a problem with pretty much all cross cultural work, in psychology, linguistics, anthropology, or whatever discipline. The authors of the paper present some fairly strong evidence that language differences in the test weren't a problem. First, the Big Five Inventory has been extensively tested in both languages for reliability, and second, while there were overall differences between the tests on the two languages (the substance of the paper), differences on individual questions were minimal, indicating substantial test-retest reliability across languages. Of course, it's still not clear whether the small differences that did emerge might be artifacts of the differences in the tests themselves, and that's why I say at the beginning that I'm not sure just how seriously to take this.

Chris said...

OK, one more comment on your question, Anon: while it is possible to argue that differences in the test are responsible for differences in scores across the two languages, the fact that the differences on all but one of the dimensions were in the predicted directions places the burden on the "differences in the test" alternative hypothesis to come up with evidence that test differences will yield that pattern of results.

Anonymous said...

When I wrote The Language Imperative, I collected questionnaires from more than 100 multilinguals around the world, and one question I asked them was whether they felt that they were in any way a different person when they switched from one language to another. The responses were uniformly of two types: "Yes, of course! What a ridiculous question!" and "No, of course not! What a ridiculous question!"

Suzette Haden Elgin

Chris said...

Haha... that sounds about right.

Anonymous said...

Hi
YOur bolg is very nice. I have really learnt a lot from this blog thanks

Anonymous said...

Nice blog dear i have really learn a lot from this blog thanks.

wow power leveling said...

cheap wow power leveling buy wow gold cheapest wow power leveling CHEAP wow gold BUY power leveling CHEAPEST wow powerleveling
wow goldwow goldwow goldwow goldweiwei

Cheryl said...

It is right to say that the language primes the culture that goes with it, but I think the main problem is not bilingual, but which language has brought you up.
Pimsleur French and Pimsleur Chinese