Here’s another spring-cleaning post. I originally wrote this back in March, shortly after Malcolm Gladwell came to talk at Duke; I no longer remember why I didn’t post it at the time, except that I never finished writing it. So here it is, completed & submitted for your edification.

Since reading about the idea of transactive memory in The Tipping Point, I’ve tracked down the article that Gladwell cites on the subject:

Wegner, D. M., Raymond, P., Erber, R. (1991). Transactive Memory in Close Relationships. Journal of Personality and Social Psychology, 61 (6), 923-929.

Wegner conveniently has a copy of this paper on his site, plus a few other papers on the topic.

The findings of this paper are that couples have better recall of factual data that had been categorized by the researchers than pairs of strangers had, but only when the couple was allowed to remember the data “naturally.” When the researchers told the couple how to remember this data – that is, when the researchers said, “for this experiment, you’re the expert in such-and-such a category” – the pairs of strangers did better. They cite another study where individuals experienced memory recall problems similar to what was experienced by the couples:

These complications resemble the difficulties that occur when a new knowledge organization system is imposed on an individual who has previously been using another system (p. 924).

So this seems to imply that a similar mechanism is at work in individual memory and in transactive memory. Which is weird, when you think about it: the same memory mechanism is at work both in your own head and collectively between your and other people’s heads. They also make the comparison between transactive memory and distributed computer memory:

One way to understand such transactive memory systems is by analogy to memory sharing in computer systems. … When computers have physically separate memory systems, they can be programmed to share memory through the creation of a directory within each isolated memory system that contains an abbreviated record of the contents of other memory systems.

Physically isolated memory systems can be accessed reliably from the different computer processors linked to each system to the degree that each memory system contains an up-to-date record of the organization and general contents of the other systems. The analogy to humans is simple: If each person learns in some general way what the other person may know in detail, the two can share the detailed memories enjoyed by both. The development of a transactive memory in the pair, then, involves the communication and updating of information each has about the areas of the other’s knowledge.

So, in essence, you’re indexing & abstracting the contents of another person’s head. Their head contains fulltext, as it were, while yours contains keywords.

I think the idea of transactive memory is cool. Maybe it’s because I’ve always thought that Yvonne and I have a really efficient transactive memory. (But does she think so? 🙂 ) Maybe it’s because I’ve always thought that the idea of collective cognition is groovy, a la Karl Weick:

Weick, K. E., & Roberts, K. H. (1993). Collective Mind in Organizations: Heedful Interrelating on Flight Decks. Administrative Science Quarterly, 38(3), 357-381.

Also, I suppose, a la the Borg.

One big limitation of Wegner’s study is that they studied college student couples, so none of them had been together for longer than about a year & a half. I’d like to see this study replicated with long-term couples, partners of 10 years or more, so that they’ve had time for a really detailed transactive memory system to develop. However, Wegner found that:

Individuals from couples who had been together longer performed no differently on the memory task than did individuals who had been together a shorter time. … The length of the relationship did not moderate the observed effect of assignment. It may be that some minimal level of transactive memory – at least for the domains of memory items used in this study – is achieved rather rapidly in a relationship, and that within the limited range of closeness available in this sample strong influences of closeness would not be observed.

But I hypothesize that this finding, as they suggest, is in fact a result of the narrow range of lengths of time that couples had been together. Other types of relationships that might be interestng to investigate in studies like this are long-term relationships other than partners: bosses & assistants, officemates, colleagues, etc. RQ: What, if any, patterns exist in the areas of knowledge expertise that individuals in different types of relationships develop?