Mixing Memory

Research on cultural differences between East Asians (Japanese, Korean, and Chinese in most studies) and Western Europeans/Americans (mostly Americans) have shown, among other things, that westerners tend to reason analytically, while East Asians tend to reason more holistically. This means, among other things, that westerners tend to pay more attention to focal objects at the expense of context, while East Asians pay more attention to context1. It also means that, when attributing causes to events, westerners generally focus on a few causes directly related to the event, while East Asians focus on the wider context, and a larger number of causal influences2. But what about consequences? Are their cultural differences in the way East Asians and westerners reason about the consequences of events?

The first attempt to address this question is described in a recent paper by William Maddux and Masaki Yuki3. The paper was inspired by an anecdote that Maddux and Yuki describe at the beginning of the paper, and which I hadn’t heard about. Here’s their description:

In April 2004, three Japanese citizens–two freelance journalists and an aid worker–were taken hostage in Iraq. Their captors threatened to execute them if the Japanese government did not withdraw its troops from Iraq. Although the Japanese government refused to bow to the captors’ demands, a week later the hostages were released unharmed and returned safely to Japan. Surprisingly, however, the hostages were met with severe criticism from the Japanese public upon their return home. The hostages were vilified as being “selfish” and “reckless.” A Japanese official in the foreign ministry indicated that “when it comes to a matter of safety and life, I would like them to be aware of the basic idea of personal responsibility.” The Japanese government indicated that they would bill the hostages for the financial costs incurred in releasing them. Ultimately, the hostages had to seek psychiatric help for dealing with the stress of the public’s negative reaction toward them, a level of stress that they said was more intense than it had been when their lives were being threatened in Iraq. They eventually issued a public apology for having “caused trouble” (Onishi, 2004).

I don’t think I need to tell you that the reaction of the Japanese government and public was very different from that of the U.S. government and public when U.S. civilians have been kidnapped in Iraq and elsewhere. There are probably many reasons for the different reactions, but Maddux and Yuki hypothesize that cultural differences in the way we reason about consequences might have played a role.

To test this hypothesis, they conducted four largely exploratory studies. In the first study, twenty-three Asian American and 18 European American participants were shown this picture of a pool shot:


They were then asked how much the shot in the picture would affect the player who would take the next shot, the player who would take the third shot after the pictured shot, the player who took the sixth shot after this one, and how much it would affect the outcome of the game. Here is their graph presenting the results:


As the graph shows, European American participants thought the shot in the picture would affect the next shot more than the Asian American participants did, but the pattern quickly reversed. Asian American participants rated the impact of that shot on the third shot as slightly greater than the European American participants, and their ratings of the impact of the pictured shot on the sixth shot were almost twice as high as those of European American participants for the sixth shot. The difference was even greater for the outcome of the game. This study provides initial evidence that East Asians think about consequences differently from westerners. Specifically, they tend to think about the consequences more globally and long-term, while westerners are more narrowly focused, as they are with causes.

In the second study, Maddux and Yuki showed 43 Japanese participants and 35 American participants a photo of a wilderness scene, and asked them to list the consequences of turning the area in the photo into a national park. Their answers were then coded as direct (e.g., “environmental improvement in the immediate area”) or indirect (e.g., answers about the “legal and political consequences”). The prediction, based on the results of the first study, was that the American participants would list a greater proportion of direct consequences, and the Japanese participants would list a greater proportion of indirect consequences. That’s what they found. Sixty-two percent of the consequences listed by Americans were rated as direct by naive coders (i.e., coders who didn’t know the purpose of the study), and 54% of the consequences listed by Japanese participants were rated as indirect. Once again, western participants seemed to be taking a narrower view of the potential consequences than East Asians.

In the third study, 72 Japanese and 90 American participants read the following scenario:

You are the president of a large company. Your company is having major financial difficulties and you decide you must lay off 15% of your employees in order to try to make the company profitable again. You meet with all the high-level managers to decide which employees are the least essential to the company and you decide to fire these nonessential employees. In addition, you decide to cut all salaries, including your own, by 15%. You hope that these measures will make the company profitable again.

Participants were then asked how many people they thought would be affected by the layoffs and pay cuts, and about their level of responsibility in each of the consequences of the layoffs and pay cuts (the firings, pay cuts, and the effects on the families of those they fired, along with an increase in crime in the area a year later). On each of these questions, they rated their responsibility on a 1-5 scale, with 5 indicating “completely responsible.” Here’s their graph of the results:


As you can see, for cutting their own salary, American participants rated themselves as more responsible than Japanese participants, and their ratings of their responsibility for the other paycuts were equal. For the other questions, however, the Japanese participants rated themselves as more responsible than the American participants did. On average, the Japanese participants also thought that almost 3 times as many people would be affected by the layoffs and pay cuts than the American participants.

The final study was similar to the third, but with a different scenario:

It is Monday morning and you are driving to school on the city’s largest and busiest road. You are the president of the student government and you are in a hurry to make it on time for an important meeting. The student government is meeting to vote on several issues of interest your school, and by the rules of the student government, they cannot vote unless you are present. You glance down to review your notes for the meeting and as soon as you do the car in front of you brakes to avoid an animal running across the road. You look up again, notice you are about to hit the car, but you can’t put on the brakes in time. With a loud crash, your car slams to a stop as you rear-end the car in front of you.

Once again, for direct consequences of the events in the scenario, American participants ratings of their responsibility were equal or greater to those of Japanese participants, and the Japanese participants thought more people would be affected than the American participants. The Japanese participants also reported that they would feel worse about the negative consequences of the events than American participants did. The results of the third and fourth studies therefore provide further support that East Asians take a more global perspective on the consequences of events than westerners (Americans) do.

It’s pretty easy to see how these results could be a consequence of holistic vs. analytical reasoning. Since holistic reasoning emphasizes context, peripheral consequences will be more salient for people who tend to reason more holistically. So the results of these four studies aren’t really surprising, though they do provide another demonstration of the implications of East Asian holistic thinking vs. western analytical thinking. Unfortunately, they don’t really provide any clues about the reasons why East Asians and westerners have different cognitive styles. So far, there’s been very little research designed to answer that question (see these two posts for descriptions of preliminary attempts to test two different hypotheses). I have to admit, I’m starting to get frustrated with this line of research, even when I find the results interesting (as I do in the case of the studies I just described). I know that interesting demonstrations are sexier than studies that test causal hypotheses (they definitely make for better blog posts), but it’s now pretty clear that East Asians and westerners reason differently, so it’s time to get our hands dirty and figure out why. My only hope is that each new demonstration will provide new avenues for testing causal hypotheses.

1See e.g., Masuda, T., & Nisbett, R. E. (2001). Attending holistically versus analytically: Comparing the context sensitivity of Japanese and Americans. Journal of Personality and Social Psychology, 81, 992-934; Chua, H.F., Boland, J.E., & Nisbett, R.E. (2005). Cultural variation in eye movements during scene perception. Proceedings of the National Academy of Sciences, 102(35), 12629-12633.
2Choi, I., Dalal, R., Kim-Prieto, C., & Park, H. (2003). Culture and judgment of causal relevance. Journal of Personality and Social Psychology, 84, 46-59.
3Maddux, W.W., & Yuki, M. (2006). The “ripple effect”: Cultural differences in perceptions of the consequences of events. Personality and Social Psychology Bulletin, 32(5), 669-683.


  1. #1 ME
    November 21, 2006

    This sounds similar to the differences in viewpoints of conservatives and liberals. Perhaps it’s possible that there are two competing systems in the brain for processing causality: one explores causality on a local level, for example the direct results of one event, and the other system explores causality on a holistic level. People would gradually learn to emphasize one system versus the other based on their experiences.

    This implies that anyone can reason from either perspective, so this could be tested by somehow holistically priming westerners, perhaps by showing them environmentalist films, and then performing the experiments in your post. In the case of easterners, the same result could be accomplished by showing them action movies, for instance, where people are shown disregarding holistic effects.

  2. #2 Joe Viscomi
    November 21, 2006

    I think this is an interesting study but I do have a minor criticism. While I do believe the difference are “real”, the third example does not support the hypothesis, and, since the fourth example is like the third and we see no data, this, in my mind, also calls into question how well the data supports the hypothesis.

    My concern is with the scale; first, a five-point scale does not offer a lot of discrimination for the respondent . In all instances the differences are less than one point and the differences in 4 of 5 instances is less than half a point. We have no idea (outside of trend) if the differences are “real” or due to random chance.
    Even if the sample is a valid sample and statistical testing indicates significance, I would ask whether or not it demonstrates practical significance.

    I believe the hypothesis is reasonable but I disagree that the data for these particular instances support that hypothesis.

  3. #3 mark
    November 21, 2006

    Well, the hypothesis is merely that cultural differences explain the way two different groups reason about consequences. My nickel psychological analysis shows that a worship of one’s ancesters, and also, perhaps, “island mentality” might have something to do with it. (I hope, Chris, that you don’t mind non-specialists commenting here).

  4. #4 Koray
    November 21, 2006

    Hmm, should one think like a westerner or east asian in order to tackle this problem?

  5. #5 Joshua
    November 21, 2006

    I think one should think how one is most comfortable thinking. ;)

    ME’s hypothesis about different kinds of reasoning being emphasised differently in the “eastern” versus “western” cultures is an interesting one. What I wonder, though, is if it isn’t really the same kind of analysis but simply with a different emphasis? Or, put another way, it’s two slightly different inputs to the same system.

    How would we distinguish between those cases? The priming experiment is a good one, but would the expected results be different between a two-system versus a one-system model? I suspect not. Is this a case where some kind of brain activity mapping could come in handy? If there are two distinct systems for reasoning, one holistic and the other analytic, one would expect to see different areas activated in the brain when a holistic thinker and analytic thinker attack the same problem. You might then back that up with a priming experiment that gets the holistic thinker’s brain to light up with a similar pattern to the analytic thinker’s and vice versa. On the other hand, if it’s the same system with different inputs, you would expect more or less the same pattern in all the subjects, regardless of thinking style.

  6. #6 Chris
    November 21, 2006

    ME, in fact, you can induce people to think one way or the other. One of the posts I linked discusses research that does this:


  7. #7 ivan
    November 21, 2006

    maybe it would be interesting to know the right vs. left handedness as well as the right vs. left hemisphere dominance in the participants.