The surprising effect of culture on vision

A number of studies have found cultural differences in visual cognition. For example, Takahiko Masuda and Richard Nisbett found that when Americans watch a short video clip of an underwater scene, they tend to recall the items in the foreground: the fish. Japanese people watching the same clip recall the items in the background: rocks, plants, and their relationship to one another.

A team led by Shinobu Kitayama showed people a frame with a line drawn inside. When asked to duplicate the line in a different-sized frame, Americans were better at drawing it the same size, despite the frame, while Japanese were better at scaling the line to correspond to the size of the new frame.

But why should such basic elements of perception vary between cultures? Most scholars have argued that social and learning differences between cultures are responsible for the difference, but Yuri Miyamoto, working with Nisbett and Masuda, wanted to explore another possibility: that these perceptual differences are due to the different physical environments in America and Japan.

They began by taking photos around three types of locations in three different-sized cities in each of America and Japan. They randomly chose post offices, hotels, and elementary schools in New York, Ann Arbor, and Chelsea, and Tokyo, Hikone, and Torahime. They stood at the front entrance and one block behind each of these sites and took photos facing four different directions, for a total of 976 photos. Here are some of the photos they used:

i-391e04fa9d1a706dff0c43e60cfe1c0d-city.jpg

Then American and East Asian students rated each photo along four different dimensions: ambiguity, number of objects, visibility, and organization. Japanese cities were rated as significantly more complex and ambiguous than American cities. East Asians tended to rate all the scenes as less complex and ambiguous than Americans. The researchers also analyzed the photos using a computer, and again found that Japanese scenes were more complex than American scenes. So it's certainly possible that being raised in a more complex environment would cause Japanese people to perceive scenes differently than Americans.

Next Miyamoto's team wanted to see if they could use these photos to influence attention. They picked a subset of the photos and told a new set of viewers that they were simply evaluating a set of pictures for a different experiment to be conducted in the future. The viewers were divided into two groups, one which rated 95 pictures from Japan, and one which rated 95 pictures from America. Next they participated in a "different" study, which asked them to look at animated clips of "neutral" scenes: construction sites or airports which could either be Japanese or American. It was a change blindness task, much like the one we discuss here. Viewers saw one version of a scene for 20 seconds, then saw another scene, in which several of the objects had been changed. A previous study by Nisbett and Masuda had found that Japanese people tended to detect more changes in the background, while Americans noticed more changes in the foreground. Here are the results of the new study, for Americans:

i-5b4c8598e2910adaf8d009c15cbcfe51-citya.gif

Americans, after viewing 95 scenes from Japan, notice significantly more background details changing in the animations than Americans who rated the 95 scenes from America.

The same task was repeated in Japan, with Japanese participants. Here are those results:

i-70ed80b9f5126876d4cfca1a95c17da1-cityj.gif

Though overall Japanese viewers noticed more background changes than Americans, they showed the same pattern: the group that had rated scenes from Japan noticed more background changes than the group that rated American scenes.

So even within just the few minutes of an experimental task, the visual experience of a person's physical environment can affect how he or she attends to scenes. Given that Japanese and Americans are immersed in these different environments for their entire lives, perhaps it is not surprising that they notice different aspects of scenes. Miyamoto et al. also point out that it is cultural differences which gave rise to the differences in physical environment, so the question of whether it is primarily culture or physical environment which affects perception and attention is an open one.

Miyamoto, Y., Nisbett, R.E., & Masuda, T. (2006). Culture and the physical environment: Holistic versus analytic perceptual affordances. Psychological Science, 17(2), 113-119.

More like this

This study was previously discussed at Mixing Memory. I'll repeat my comment from there. If the authors want to argue that the differences in complexity and ambiguity that they identify in Japanese and American scenes have a causal role in producing the differences in perception, then they should isolate this variable and see what happens when they prime people with photos that differ in complexity-ambiguity but are otherwise as similar as possible (most importantly, from the same country). Otherwise, we have no particular reason to think that it is the differences in visual complexity and ambiguity, rather than some other difference between the Japanese and American scenes (such as the cultural content), that is influencing visual attention on the next task.

If I remember correctly Melville Herskovits (anthropologist) wrote about differences in viewing optical illusions between West Africans and North Americans which he attributed to differences in the built environment. I have not been able to locate a good bibliography of Herskovits work to track it down, but as he died in the late 60s the paper is likely to be from the 50s (or earlier).

By Stan Jones (not verified) on 26 Apr 2006 #permalink

Why is this cultural and not locational? Why didn't the researchers test Japanese raised in the US and Americans raised in Japan? To identify this as a cultural difference identified by the words Japanese and American make this easily transferrable to race, and confuses the issue. But then, "cultural differences" is the hot phrase right now so that results in funding and publication.

By John Tenny (not verified) on 06 May 2006 #permalink