My Favorite Experiments: Bransford and Johnson

I've been busy as hell, so I haven't had much time or energy to post anything lately. But I had an idea today that I thought I'd try out. There are a bunch of experiments that I really like for various reasons, and because I really like them and have described them so many times, I can write ten pages about them in my sleep (and given the fact that I'm so tired, I may well be doing that by the end of this post). So I'm going to try to post on one of these every now and then until things slow down a bit here over the break. I'll try to give a bit of context and explain why I like the studies so much. And since several of the experiments I really like are related, maybe a coherent picture will emerge at some point. I'm going to start with Bransford and Johnson (1972, 1).

OK, so for a really long time -- like two millennia -- memories had been conceived of as copies, or in more recent parlance, traces of the experiences they represented. You can find something like this view in Aristotle, and especially in the British Empiricists. Here's a pretty representative quote, from Hume's A Treatise on Human Nature(2):

We find by experience, that when any impression bas been present with the mind, it again makes its appearance there as an idea; and this it may do after two different ways: either when in its new appearance it retains a considerable degree of its first vivacity, and is somewhat intermediate betwixt an impression and an idea: or when it entirely loses that vivacity, and is a perfect idea. The faculty, by which we repeat our impressions in the first manner, is called the MEMORY, and the other the IMAGINATION. 'Tis evident at first sight, that the ideas of the memory are much more lively and strong than those of the imagination, and that the former faculty paints its objects in more distinct colours, than any which are employ'd by the latter. When we remember any past event, the idea of it flows in upon the mind in a forcible manner; whereas in the imagination the perception is faint and languid, and cannot without difficulty be preserv'd by the mind steddy and uniform for any considerable time.

Here, when our representation of an experience retains all the good stuff from the experience, what we'd today call episodic information, and Hume calls colour and vivacity, it's a memory. When it loses this stuff, it becomes a "perfect idea," or what we might today refer to as concepts, which is the property of the imagination, and entirely separate from memory. Memories, then, are just copies of the experience, and the perfect ideas are something entirely different (of a different species, Hume says in the next sentence).

This view of memory was still pretty dominant when empirical psychology was born in the 19th century. Ebbinghaus, for example, with all his nonsense syllables, held such a view. For him, memory was a matter of traces that decayed (forgetting), or were distorted, blocked, or overwritten (interference). Behaviorists were even worse. They pretty much did away with the concept of "memory" altogether, though they still held some sort of associationist conception that looked a lot like memory traces.

Then came the cognitive revolution in the middle of the last century, and bunch of young whippersnappers were eager to change everything, so they latched onto an idea that had been simmering underneath the smothering layer of behaviorism and empiricism that had dominated empirical psychology for what seemed like forever. The idea, which you can trace to people like Wundt and Titchener (and my buddy William James), and which was described most explicitly (but was subsequently ignored, at least in the U.S.) by F.C. Bartlett, was that memory and Hume's imagination, or concepts, weren't really separate things. Under this view, memory wasn't about copies or traces, but conceptual representations, sometimes referred to as schemas (though no one knows what the hell a "schema" is). Retrieving a memory was no longer a process of picking out a whole copy of an experience from some storage bin in memory, but of reconstructing the experience using the schema as a guide.

This view sparked a ton, and I mean a ton of research in the 1950s, 60s, and 70s, most of which centered around a few key principles: selection, abstraction, integration, and interpretation. Basically, this view posits that when we encode a new experience into memory, we first activate a schema, which selects which aspects of the experience we encode. These schemas are abstract, because they're used to encode a bunch of related experiences that may have different specific properties but the same overall structure. So when schemas are selecting information to encode, they get rid of the specific details of the new experience. Thus, the new experience is integrated into the schema, and structural details that are new and relevant to the schema can be added to it, allowing the schema to handle an even wider range of experiences. Finally, the experience is interpreted through the schema, so that information in the schema but not immediately apparent in the experience can be inferred.

It might help to give you an example. Say you walk into McDonald's. You automatically activate your McDonald's schema, which tells you things like how you order (at the counter), that you have to pick up your own food 'cause no one's going to bring it to you, that they give you a cup and you get your own drink from the soda fountain (do they still call them soda fountains?), etc. If you're later asked to remember your trip to McDonald's, you'll likely remember all of these things, and forget all the incidental details of the visit. That's selection and abstraction. If you get in and find that they no longer give you a cup so that you can pour your own drink, but instead fill your cup for you, you'll probably stick that into your schema (especially if it happens on a couple visits). That's integration. If instead you walk up to the counter, order your food, and then receive your food without getting a cup, you'll ask for one, 'cause you know you're supposed to get a cup so you can get your own drink. That's interpretation.

It's pretty simple, right? It's simple, but it's a really powerful theory, and it can have some profound implications -- false memories, for example. As I said before, there was a ton of research done in the first twenty years after the cognitive revolution testing, or at least demonstrating, different consequences of this view. I could write a blog just on that literature, and I'd have a couple years' worth of material. But I'm only going to give you one set of studies today, described by Bransford and Johnson in a 1972 paper, and primarily designed to demonstrate the selection principle, because it's one of my favorites.

In their first study, Bransord and Johnson gave their participants this paragraph:

If the balloons popped, the sound wouldn't be able to carry since everything would be too far away from the correct floor. A closed window would also prevent the sound from carrying, since most buildings tend to be well insulated. Since the whole operation depends on a steady flow of electricity, a break in the middle of the wire would also cause problems. Of course, the fellow
could shout, but the human voice is not loud enough to carry that far. An additional problem is that a string could break on the instrument. Then there could be no accompaniment to the message. It is clear that the best situation would involve less distance. Then there would be fewer potential problems. With face to face contact, the least number of things could go wrong. (p. 719)

Doesn't make much sense, does it? Some of the participants got just this paragraph, and they didn't think it made much sense either. They also couldn't remember it very well. Another group of participants saw this picture before reading the paragraph (from Bransord and Johnson's Figure 1, p. 718):


Suddenly the paragraph makes sense, right? This group of participants certainly thought so, and they were able to remember it pretty well too. Another group of participants saw the same picture after reading the paragraph, and a fourth group saw only part of it before reading the paragraph, with enough missing to make it difficult to tell what the picture was about. Neither of these groups could make much sense of the paragraph, and they didn't remember much of it either.

In this experiment, you can think of the picture as the schema. It serves to structure the information you get in the paragraph, and select what you remember about it. Without that structure, you don't having anything to pick what you should remember, so you don't remember much of anything. And the fact that the picture only worked if you gave it to the participants before they read the paragraph is important, because it suggests that schemas are doing their work at encoding, that is, when you're storing the new information.

In their subsequent experiment, participants were read a paragraph like this one:

The procedure is actually quite simple. First you arrange things into different groups. Of course, one pile may be sufficient depending on how much there is to do. If you have to go somewhere else due to lack of facilities that is the next step, otherwise you are pretty well set. It is important not to overdo things. That is, it is better to do too few things at once than too many. In the short run this may not seem important bu complications can easily arise. A mistake can be expensive as well. At first the whole procedure will seem complicated.
Soon, however, it will become just another facet of life. It is difficult to foresee any end to the
necessity for this task in the immediate future, but then one never can tell, After the procedure is completed one arranges the materials into different groups again. Then they can be put into their appropriate places. Eventually they will be used once more and the whole cycle will then have to be repeated. However, that is part of life. (p. 722)

One third of the participants just heard the paragraph, and as in the last experiment, they didn't think it made much sense and couldn't remember jack about it. Another third of the participants were told that the paragraph was about doing laundry before they heard it. These participants thought it made perfect sense, and remembered a lot about it. A third set learned the topic of the paragraph after they'd heard the paragraph. These participants didn't think it made much sense, and like the first group, couldn't remember much about it. In this case, the topic served to activate the relevant schema -- doing laundry -- and unlike in the first study, it was a schema that the participants already had somewhere in their heads. Just like the picture, though, the topic helped the participants to work out the structure of the paragraph so that they could make sense of it, and this allowed them to remember more about it. That's selection, baby.

So that's Bransford and Johnson, one of my favorite set of studies. Next up, either Bransford and Franks with integration or Sullin and Dooling with interpretation (I haven't decided yet), whenever I get around to it.

1Bransford, J.D., & Johnson, M.K. (1972). Contextual prerequisites for understanding: Some investigations of comprehension and recall. Journal of Verbal Learning and Verbal Behavior, 11, 717-726.
2Book I, Part III, Section III.

More like this

I have to say it's a little weird to see a paper by people one was in grad school with cited as a sort of classic. If you do the Bransford & Franks papers next, you might be interested to know that around the U of Minnesota psych department at that time (actually, in the Center for Research in Human Learning) they were known jointly as "Franksford," since they were always together. :)

Has anybody using the premises of selection, abstraction, integration, and interpretation and research been able to conclude anything on the subject of creativity ?

Those paragraphs are so generic, they could do an experiment in creating false memories by showing the participants a different picture than the one they actually saw and asking for their memories again.