epistemology and cognitive development

Cognitive developmentalists (cog devs) have, since at least the time of Piaget, taken an epistemological approach to understanding children. By epistemological, I mean that their investigations have been concerned with determining things like what children of different ages know, and the progression of things kids know in a given area, and so forth. (obviously behaviorists are not really included in this). This approach has made substantial contributions to a large extent because of how different children are than adults in terms of their thinking. Familiar types of activities that seem simple and obvious to adults are approached in completely different ways by kids. Inabilities in a certain area have been attributed to children not understanding things about that particular area.

Despite the simple production of "results" using this approach, researchers have done little more than open up previously unknown areas of investigation. Empirical results of (e.g.) children's numerical thinking have made it very obvious that children think about numbers and counting differently than adults do. Yet, it has been virtually impossible to move from negative statements about knowledge taht can be easily backed up by research, to positive statements that can be backed up and conclusively agreed upon. Instead, what ends up happening is that researchers make assumptions about the epistemological implications of a given behavior (what a certain behavior indicates that a kid knows), and draws conclusions on the basis of these generally questionable assumptions.

The assumptions are questionable because it is really hard to come to any conclusion about whether or not a kid "knows" something, for several reasons. First, behaviors are not proof of a certain form of knowledge. I may be able to count because I have been taught counting by rote, or I may be able to do it and understand its function and how to do it correctly based on sound mathematical principles. The obvious reaction to this problem is to verify abilities seen in one area with abilities in another. E.g., to determine whether a child who can count a row of objects really understands counting, they could be asked tocount to comapre two sets of objects, or to pass a conservation task. Although these demonstrations may make certain conclusions increasingly plausible, the same objection that applied to the initial findings apply to these.

The second difficulty with attributing knowledge to kids is that it's not at all clear what knowledge is. I can say that I know how to drive a car, and this may be meaningful and effective for many social purposes such as whether or not I can take a shift on a long driving trip. But this just means that "my knowing something" has certain practical implications, specifically that it allows my behavior in a variety of situations to be predicted. This sense of knowledge is often useless for children whose ability to perform successfully on related tasks is often baffling and surprising. For example, researchers have repeatedly demonstrated that children who can count a set of objects are unable to give out a requested number of objects when asked. This combination of ability and inability is particularly baffling to adults who can hardly imagine doing one without doing the other. It's not at all clear what kind of "knowledge" these children would have to have to explain this constellation of abilities.

In my opinion, the concept of knowledge doesn't refer to a well defined state, but rather serves a functional purpose to make claims about the sorts of activity that might be expected from a certain person, based on past events. These predictions are grounded in the system of established regularities of human interaction, most of which are not understood.

Based on this closer look at the concept of "knowledge" (this applies to its close cousin understanding as well), what is its use in cognitive development?

I would argue that the use of epistemological terms such as "knowledge" in cognitive development results from prevailing cultural practices that are in place for dealing with thinking. We use these epistemological terms to make sense of other people's behavior and thinking because they are socially useful for these purposes. Because of the terms' usefulness in these contexts, they may have become reified, or treated as more real than they actually are. Continued use over time may have obscured the fact that epistemological terms like knowledge are representations of cognitive processes, not reflections of them. The result is that people continue to use them for purposes where they appear to have little use.

Cognitive development is the perfect example of this phenomenon of the reification of concepts leading to confusion within the field. Instead of serving to make sense of the subject matter, these concepts instead dominate the subject matter, to the extent that the actual subject of study is often ignored, and replaced with an alternate object that more closely fits the concept. This can be seen when researchers are trying to determine (e.g.) whether or not children know that numerosity is a property of all number words. The idea that children possess this "knowledge" is forefront in the investigation, which is concerned with detecting its presence in children of a certain age. Investigators assess different aspects of children's behavior, and draw conclusions about whether or not they possess the knowledge in question. All the while, it is never considered what the very idea of "having knowledge" even means for the child. It is simply assumed that the proper object of cog dev research should be to detect the supposed entities such as knowledge that exist beneath the surface, influencing (in certain contexts) children's activity.

This approach seems to be ridiculous when we reflect back on the earlier consideration of what it even means to have knowledge. Knowledge, it was concluded there, is a practical way of making sense of thinking for certain purposes. As the research example shows, however, this has become inverted. Rather than being the tool for generating understanding, knowledge has become the goal of the investigation. The incredibly ironic result is that the investigators have a thinking child in front of them during the experiment, and yet they ignore this thinking because of their preoccupation with the concept of knowledge, which social interaction has reified into an object of study, rather than a way of representing knowledge.

What has happened here can be humorously illustrated with an example. Imagine a detective who has been hired to track down a burglar, and has been given a picture of the burglar as a guide for who to look for. The detective then goes out with the picture and begins drawing other pictures of the faces he encounters, which he then compares to the original picture, in the hope of finding the thief.

This method of tracking down the criminal is analogous to the approach taken in cognitive development. Just as the detective placed too much importance on matching his subjects to the picture, so too do cognitive development researchers forget that culturally received ways of representing cognition are not reflections of cognition itself. All too often they, like the detective overlook the actual thought processes that are the real subject matter of cognitive development, preferring instead to look for something that perfectly matches a particular representation of cognition.