The foreign-language effect: Thinking in a foreign tongue reduces decision biases

American research, September 2012

The Power of “OUT” – LGBT in the WorkplaceDecision-making is influenced by the information that we systematically process, as well as our emotional system (what we value). It is the latter that can often leave us open to decision biases. So, how can we ensure that our decisions are, so far as is possible, unbiased?

This research, conducted by Professor Boaz Keysar, and graduate students Sayuri Hayakawa and Sun Gyu An from the University of Chicago, explored the influence of language on making decisions. They posed the question, “would you make the same decisions in a foreign language as you would in your native tongue?” (Keysar, Hayakawa & Sun Gyu An, 2012).

In essence, the researchers found that decisions made in a foreign language were less biased than decisions made in a native tongue. A key finding was that cognition in a foreign language provided greater emotional distance during the decision-making process. Of course, this begs the question how can this finding help individuals within a workplace setting aside from learning and communicating in a second language?


This research aimed to explore whether language impacts our ability to make decisions.


Experiment 1

Experiments 1a through to 1c tested for ‘framing’ effects (i.e. how we frame a decision). When faced with a choice between saving the lives of 200 out of 600 people, or to take a chance at saving all or none (a gain-frame scenario), this research suggested participants were more likely to prefer saving 200 lives. Yet when this decision was framed differently, to lose the lives of 400 out of 600 people or to take a chance at losing all or none (a loss-frame scenario), participants were more willing to take the risk. The differing preferences for the two identical scenarios demonstrated how emotion can influence decision making.

Using similar scenarios, participants were asked to make a choice between Option A (a certainty of people or jobs, lost or saved) or Option B (taking the ‘all or none’ chance, with unfavourable odds). Experiment 1a included 121 students from universities in Chicago, Illinois, Raleigh and North Carolina. All were native English speakers who spoke fluent Japanese. Participants were randomly assigned to either a ‘gain-frame’ or ‘loss-frame’ version of the scenario.

Similarly, experiment 1b included 144 students from Chung Nam National University in Daejeon, Korea. Participants performed the task in either their native tongue (Korean), or in a foreign language (English). Experiment 1c included 103 native speakers of English, who were studying in Paris, France. In experiment 1d, the researchers completed a control experiment with 84 University of Chicago students, to ensure that the previous experiments were not due to random selection by participants.

Experiment 2

In Experiment 2, the researchers tested ‘loss aversion’ namely a mental framework in which an anticipated potential loss outweighs the positive impact of an identical potential gain. In this experiment, 146 native Korean speakers (with English as a foreign language) were presented with 18 equal-odds bets; all with a positive outcome (for example, lose $119 or win $170). The odds of the bet ranged from an unattractive 9/10 loss-to-gain ratio to a highly attractive 1/10 ratio. Participants made a decision on whether they would accept each bet.

Experiment 3

Fifty four students from the University of Chicago, speaking Spanish as a foreign language, had to choose whether to accept a bet. However, in this case they were provided with money that they could keep after the experiment. In each round, participants could choose to bet $1 and receive an additional $1.50 if they were successful, or lose their original $1 if they were unsuccessful. They had the option of declining to bid and keeping their money.


The researchers found that in Experiment 1, participants completing the task in a foreign language appeared less likely to be influenced by how a decision was framed. In Experiment 2, participants completing the task in a foreign language appeared to be less affected by the fear of loss and more willing to accept bets. In Experiment 3, participants who performed the task in a foreign language took the bets more often than those who performed the task in their native language.

Overall, the three experiments demonstrated that we are less likely to be affected by decision biases (namely framing effects and loss aversion) when making choices in a foreign language.

Experiment 1 findings

When tested in their native language, 77 per cent of participants who answered the gain-frame problem chose Option A, to ensure that a certain amount of people would be saved (either lives or jobs, depending on the scenario). However, in the loss-frame version, only 47 per cent of participants selected the same answer. Interestingly, this 30 per cent gap, which demonstrates the effects of framing, disappeared when participants completed the task in a foreign language. Experiments 1b, 1c and 1d replicated the results of experiment 1a.

Experiment 2 findings

In the betting experiment, language affected choice when the stakes were higher. Although participants’ willingness to take a large bet increased as the odds became more favourable, participants were consistently more willing to bet in a foreign language (English) than in their native tongue (Korean).

Experiment 3 findings

The participants who completed the task in a foreign language (Spanish) accepted more bets than native English-speaking students. This suggested that people take more bets in a foreign language because they are less affected by loss aversion.


These findings indicate that it is possible to reduce certain decision biases. Framing effects and loss aversion can be reduced by distancing the decisions that we make from our emotional reactions. This research has shown that speaking in a foreign language is one way that we can achieve this. The question is, how these findings can be applied within the workplace.   One obvious implication is to ensure that teams comprise native and non-native speakers, and that weight is given to the views of non-native speakers in decision-making contexts to leverage their ability to make more rational, and less emotionally biased, decisions.

To read the full article, see Keysar, B., Hayakawa, S.L., An, S.G. (2012) “The Foreign-Language Effect: Thinking in a Foreign Tongue Reduces Decision Biases” Psychological Science, Vol 23, No. 6 (2012), pp. 661-668.