Monday, May 19, 2008

The spirit of Piaget

Simon Ingsreview in the Sunday Telegraph made some excellent points about the difficulty of writing about toddlers' minds:
This is the problem: babies are bags of bits. They acquire and mislay abilities at a dizzying rate. Development is not a series of lessons learned. Some babyish 'abilities', such as stepping and swimming, are primitive reflexes that vanish shortly after birth, swamped beneath a welter of fumbling experiments in gross motor control. Babies actually forget how they originally recognised faces, and reacquire the skill using a different part of the brain. It's frustrating: toddlers get up to all sorts of new tricks, but do not seem to grasp them in any obvious or convenient order. At around eight months they learn how to use thumb and forefinger to pick up an object - yet they don't yet know how to let go.
Here's how I make a similar point in the book: 
Children’s minds are not like computers which, armed with a software upgrade or new memory chip, deliver enhanced performance across the board. They are assemblies of disparate skills and quasi-theoretical knowledge systems which rattle along, helping each other out when the need arises. Sometimes children use a theory developed in one domain, such as social reasoning, to solve problems in another, such as physical causation.
Ings is right, then, to say that babies are 'bags of bits'. You can't expect to find some general principle that explains why they come to do what they do when they do, and then sometimes lose the capacity again temporarily. To put it another way, we can't hope to have a general theory of development which explains all the different things we need to explain. Far more likely is that we will be stuck with a mishmash of disparate theorylets, each restricted to a particular narrow domain. Got a theory for how toddlers come to represent the intentions of others, or for how babies use visual feedback in correcting their reaching movements? Fine! Just don't expect it to explain anything else. 

This is how the spirit of Piaget has most tested modern developmental psychologists. Here is Alison Gopnik
Once upon a time, thanks to Jean Piaget, the field of cognitive development had a coherent, interesting, testable, and widely accepted theory. Now, alas, we are back in the preparadigmatic boat with our colleagues in the rest of psychology, with theory fragments, almost-theories, and pseudotheories bobbing about around us. 
We had a grand developmental theory, Gopnik is saying, but it didn't last. It couldn't quite match up to the sheer complexity of children's minds. Instead, we need to find different theoretical tools for each explanatory task, and be glad that we can explain, in separate and circumscribed ways, at least some of the varied abilities that babies and toddlers display. 

This is actually a slightly different point to the one that Ings is making. Fragmentation of abilities does not have to entail fragmentation of theory. As the modern proponents of dynamic systems theory have shown, you can have a system based on some simple underlying principles which nevertheless behaves in an apparently chaotic and back-to-front way. Piaget actually had no problem with the idea that children sometimes have to go backwards to go forwards. In fact, he and his followers showed that development is littered with U-shaped curves, as they have become known. It is a story that is familiar to developmental neuroscientists such as Mark Johnson, who (as in the example given by Ings) have shown that backward steps in development are often simply the result of one brain system taking over a function previously performed by another. 

There is another sense, Ings notes, in which I owe a debt to the great psychologist: 
The spirit of Jean Piaget - the Swiss psychologist who turned a father's fascination with his offspring into a fertile natural science - hovers over this book. When Fernyhough needs to sum up an idea about development quickly and accurately, he looks to his daughter, and ... simply tells us what he sees: the look of comic concentration with which Athena registers the effects of an action; the surreal cack-handedness of her first jokes.
I certainly don't deserve the comparison with Piaget, but I did want the book to be driven by observation rather than theory or prescription


  1. I find dynamic systems theory (DST) really hard to understand! I remember that, in an essay, I argued that DST and connectionism allow us to build models to describe and explain development, but that they are not theories in the same sense as Piaget's - they do not make predictions as such - and so are not candidates for a new grand developmental theory. However, my comments were mainly aimed at connectionism as I have a better grasp of that! I'm not sure if that argument really applies so much to DST. What would you say? Anyway, perhaps that's a distinction that is not very important ... Maybe a better question would be: Do you think DST is the way forward?


  2. Charles FernyhoughJune 2, 2008 at 8:07 AM

    I agree that DST is probably a metatheory - a method for constructing theories - rather than a theory in itself. But perhaps it gives us a way of thinking that would allow the construction of grand theories proper. van Geert's use of the method would seem to be a good example of that. I'd love to have the mathematical expertise to do this kind of work myself - but that might have to wait for the future!