Blog of Sara Jakša

Why not to Trust just one Study (Language and Decision Making)

I like the Chen's study about connection between the language and decision making. His hypothesis was basically, that whenever people need to use future, affects the time discounting. So the language that the people use, would have some effect on the decisions about savings, and health related behaviors, like smoking and exercise. The language which need to use the future (like Slovenian, English, and Italian) would have a bigger time discounting, so they would be less likely to save and more likely to smoke and have unprotected sex than language which don't need to use future tense to talk about the future events (German, and Finnish).

And I was quite quick to agree with it. I mean, I am an economist, and he made a pretty good argument, why any other variable would not be responsible for these differences from the economics point of view. I mean, I have read the papers that were talking about the language differences in people fluent in multiple language, depending on which language they used: foreign or native. There were differences in moral decision making, like the trolley's problem and economics decision making. There are also neurological differences in theory of mind tasks, depending on the langauge So I guess I was pre-primed into believing this.

But I guess things can not be done that quickly. For example, this article tries to expand on the original one by including the language connectivness. They wanted to see, if the connection between language, for example the same historical origin or the mixture between the neighboring language, had an effect. What was interesting in there, that they made multiple statistical tests. Some proved the above hypothesis and some did not. But on the end, they still concluded, that language might not have this effect, as least on the saving behavior, which is the only one they tested.

I, as an economist, with almost zero linguistic training, would never have though about testing for transfer between the language that co-resided together spatially. Even though I noticed some of this, for example the Slovenian and Hungarian words for forth and fifth day of the week are basically the same, even though these language do not belong together in the same family (or however the indoeuropean level is called). Even the current cognitive science training did not prepare me for this.

There is just so much knowledge involved in the evaluation of the science, that it is hard to know everything by oneself. That is why I still recommend reading additional articles in the same sphere. This way, there might be different scientists, with different prior knowledge, having different outlook on the problem. And that way, there is less chance of there being a big hole in the knowledge. Though there is most likely still going to be, but ever step into better decision making for humanity counts.