Know thyself, yes – but how?

This post was first published on the Alphapsy blog in 2006.

The importance of self-knowledge has often been emphasized, from traditional lore to New Age gurus. However, there may be very different ways to know thyself. Two of the most important aspects of self-knowledge are autobiographical memory and self-concept and it has been repeatedly shown that these aspects of self-knowledge display wide cultural variations. A recent paper reviews what we know about the developmental roots of these differences and it illustrates nicely some of the more recent and interesting trends in cross-cultural psychology.



Qi Wang has analyzed the answers given by hundreds of children and adults, Easterners and Westerners to questions like “how would you describe yourself” or “tell me about a time you did something fun”. He reviews his (and other’s) findings in the last volume of Current Directions in Psychological Science (september 2006 – editor's note).



One of the classic observation is that when asked to describe themselves, Easterners will tend to focus more on interpersonal attributes (“I love my wife”) and Westerners on personal attributes (“I am a funny person”). Wang has shown that such trends are observable from very early on: you can already find them in the transcripts of 3-year-olds. At this age, another trend is already apparent: Westerners are able to access more distant and more detailed very long term memory (in the case of 3-year-old ‘very’ long term may be an overstatement, but the effect is there anyway). Such an interest in the developmental roots of a cross-cultural difference is still too rare, and it is clearly worth mentioning (see this paper for one of the exceptions).

Two other aspects of the studies conducted in this domain are relevant too. First, it can be noted that the cross-cultural differences can be seen as being ‘continuous’ with inter-individual differences inside a culture. For example, Wang has found that among 3-year-old the tendency to focus on personal aspects of the self predicts the amount of detailed event recalled, and this independently of culture. It means that if you want to know if a child will remember a lot of detailed events, it is better to know if she tends to focus on personal aspects of the self than knowing than if she is Chinese or American. It is important to remember that the cross-cultural differences are often differences of means: it’s not all the Easterners on the one hand and all the Westerners on the other. On nearly every dimension, a lot of Easterners are going to be more ‘Westerner like’ than the average Westerner, and vice et versa.

The other point that is worth emphasizing is that the differences are often more shallow that you might think. What I mean is that the differences usually observed between two populations can often be canceled or even reversed if an appropriate prime is used. I can not resist mentioning one of the most telling studies that has used this paradigm. Miyamoto and his colleagues used pictures of Japanese and American cities to prime participants before a task that measures the importance on contextual information in visual processing (when you look at a picture depicting a salient object, are you also going to look at and remember the context surrounding the object). The usual effect is that Easterners are more sensitive to contextual information than Westerners. However when primed with pictures of Japanese cities, Americans acted more like Japanese and it was the other way around for Japanese. In this case, a difference that was thought to be deep seated had been shown to be easily malleable. Similar effects are observed in the domain of self knowledge. I mentioned above that Easterners are usually able to access less distant long term memory than Westerners. This effect can be easily cancelled: Easterners who first had to complete a task of listing personal attributes were then able to remember memories as distant as those recalled by Westerners.

This kind of research bodes well for a more subtle cross-cultural psychology.

Julien Dutant commented on this post:

I can't access the paper, so the following is just to stir up some discussion:

Is it the effect that's shallow, one wonders, or rather the protocol? Children are taken into a lab, they're asked "how would you describe yourself?" and they produce an answer. Let us call "DOS" (Describe-One-Self) the mental routine, whatever it is, that they do in answering that question. I'm not saying it's modular or anything, it's just a neutral name for the set of cognitive operations they do to produce an answer to that specific question.

The main question is: is DOS an important aspect of self-knowledge, let alone human cognition in general? There are diametrically opposed possibilities here. One extreme is: DOS accesses a Concept/Theory of Self (say, a set of attributes and memories), and that Concept of Self is actually used whenever the subject takes a practical decision (about what to cook tonight as well as choosing one's spouse). Another extreme is: DOS accesses a short verbal description of oneself (a kind of mental business card) that one uses only to describe oneself in conversations, but that has no influence at all on planing actions.

Now, if it turns out that the effects are easily reversible, why describe them as "differences in the concept of self" at all rather than mere "differences in informal conversation habits" ? It seems to me that we remain completely in the dark as long as we haven't shown whether this purported differences show up in practical decisions or at least in forced/limited-choice tasks as well.

My reply:

I'm not an expert in this field at all, so here are just a few thoughts in answer to the comment by julien.
-I don't think self knowledge is an important aspect of human cognition (as, say LTM, the language module or that kind of things). the abilities that allow us to have a self knowledge have probably evolved for other purposes (ToM springs to mind)
-as for the role of the self concept in practical decisions, the question is obvisously tricky. on the one hand, the decisions we make clearly influence our self K (if I make people laugh a lot, I might deduce that I'm funny). on the other hand, self knowledge can be used as an approx. of how others see us, and since we usually want others to see us as being coherent, we might act on the basis of this self K (e.g. i've interiorized the fact that others see me as a funny guy and i will thus tend to make a lot of jokes — i might even do this quite reflexively, by trying to pick up jokes on the internet etc.)
-i'd like to conclude that i'm pretty sure it's not just informal conversation habits — and there is some evidence to support this claim (such as: even though self K is partly situation specific, you still find quite robust interpersonal differences, and, as i mentionned, it as been shown to be the basis of some decisions — i.e. it's not just the decisions i make –> self K, it can also be self K –> the decisions i make)

No comments yet