AND MERCI TO THE FRENCH TEACHER

Why introducing a problem in a foreign language overcomes major cognitive biases

We live in a time when existential challenges that accompanied prehistoric man on the African savannah have been displaced by new challenges in the form of a necessity to choose between options, whether in matters of consumption, investment or career. While our emotional system is comprehensively equipped with a number of outdated instincts that are focused on risk reduction and immediate reward (before the food runs out), as we already have seen, its main disadvantage is its utter blindness to probability and complex calculations that are needed for meeting some of the modern world’s demands. The source of a large share of our miscalculations is our confidence that our rational system is operating, whereas we are actually responding to the domineering and prompt emotional system. This phenomenon has been given the general name “cognitive biases,” and many dozens of these have been identified to date. The individuals most responsible for this are the researchers Amos Tversky and Daniel Kahneman, and the latter’s book Thinking, Fast and Slow, published in 2011, is a fascinating journey deep into the way we make decisions and a sad portrait of the circumstances under which our minds succumb to the dictates of the emotional system only to be mistaken in their decisions.

Two of the better-known biases belong to the prospect theory that Kahneman and Tversky developed in the late 1970s—“presentation bias” and “risk aversion.” According to the first, we tend to prefer one possibility to another possibility of identical statistical expectancy, merely because of the way each is presented to us. For example, if a person close to you is sick and needs surgery, which would you prefer—an operation that has a 30% mortality rate or an operation that has a 70% chance of ending in success? Risk aversion, on the other hand, reflects the lack of symmetry between the emotional pain of failure and the pleasure of success. Most of us, it turns out, are willing to pass on a gamble that stands a positive chance of profit just to avoid the pain of loss. In fact, as Kahneman and Tversky discovered in an important experiment from 1979, the potential profit has to be double the possible loss or more to offset our risk aversion.

If indeed the emotional system is responsible for many of our decision biases, it is interesting to ascertain whether making decisions while using a foreign language creates an original bypass for the decision-making circuit in a mind that does not even know there are foreign languages, the kind of bypass that manages to neutralize the emotional component in the process. A group of researchers led by Boaz Keysar of the University of Chicago picked up the research gauntlet and tried to find out whether thinking in a foreign language in fact reduces cognitive biases that may affect decision-making. The researchers drew encouragement from earlier studies that had shown that people do not react to taboo words, scolding or even words of love that are spoken in a foreign language with the same intensity as they would if spoken in their mother tongue; in other words, activation of the part of our brain that deals with processing foreign languages reduces the influence of the emotional system, hopefully allowing a more rational decision-making system to gain ascendance.

The researchers conducted six studies on three continents with more than 600 subjects who speak five different languages. In the first series of experiments the researchers presented subjects with a version of the “Asian flu problem” that Kahneman and Tversky had developed when they wanted to model presentation bias. The pair presented their subjects with an imaginary scenario in which the United States is preparing for the outbreak of an epidemic. Subjects were asked to choose between two action plans in view of the impending epidemic. One is perceived as relatively safe by ensuring the survival of a third of the patients; the other seems riskier, might save the lives of all, but if it fails, two thirds of the patients will die. Even though in statistical utility terms both alternatives are identical (only the cost/reward ratio is different), it turns out that people tend toward clear preference for what looks like a safer option when the outcome is presented in a positive light (patients will be saved), but tend to bet on the risky option when they are trying to avoid an outcome that is presented in a negative light (patients who will die). However, when Keysar and his colleagues presented the question to their subjects in a foreign language, they coped better with the influences of presentation bias and managed to spot the similarity in the two proposals.

To test the effect of language on another bias, risk aversion, each of the subjects received $15 in single dollar bills, from which s/he took one dollar for each separate round of betting (subjects were offered a total of 15 betting rounds). In each round, subjects could decide whether to keep the dollar or bet it on a coin toss that could earn them an additional $1.50 if they won, at the risk of losing the dollar if they lost the bet. When the gamble was presented in English, their mother tongue, they were more affected by their risk aversion (the fear of loss) and only 54% decided to bet, even though the bet had a positive expected value. When the process took place in Spanish, a language they had acquired, subjects bet on 71% of the occasions. Making decisions in a foreign language, the researchers concluded, reduces the emotional response in the process, and therefore reduces the possibility of bias in making the decision.

And if this is indeed the case, is it not time for the managers of our investment plans to phrase their proposals to us in a foreign language? Absolutely, but only when they themselves decide to make their investments using a foreign language of their own.