Anemone's Research Emporium

A critique of Janet Wilde Astington's
"The Child's Discovery of the Mind"

Anemone Cerridwen
August 6, 2008

Astington has provided us with a wonderful resource for understanding the theory around "theory of mind". She describes the history of the concept: who invented the term and what they meant by it, then philosophical discussions people had about what it should mean, then subsequent experiments and what they say about human cognitive development in young children.

And then she has a chapter on autism. A blood-pressure-raising stomp-on-the-book-in-a-fit-of-rage chapter on how autistic people probably lack theory of mind, and that lack of theory of mind may be the central deficit in autism. Warning! Do not read this chapter without plenty of drugs on hand. My preference is chocolate, but you may have other drugs of choice. Regardless, proceed with caution. Do, however, read everything else, merely avoiding all sentences with the word autism in them, if you want to get a good general understanding of theory of mind.

Theory of mind was a term coined by David Premack and Guy Woodruff, who wanted to know if chimpanzees could tell what someone was thinking by the context of the situation (i.e. does this person want bananas but can't reach them?). Some of the best research on human cognition has begun with research on chimpanzees. The mirror recognition test was also originally devised for research on chimps, then later applied to toddlers. (Sneak some rouge on the nose, put child or chimp in front of mirror, watch for reaction. Does the child point at the mirror, ha ha!, or investigate his own nose?)

As Astington describes, philosophers got a hold of this concept of theory of mind and juggled it back and forth for a bit. Then the philosopher Daniel Dennett decided that if people truly had theory of mind, they would understand false beliefs, not only in a naturalistic setting, but also in standardized tests. The psychologists Heinz Wimmer and Josef Perner came up with a set of standardized false belief tests, and off we go. In general, these tests show that three year olds cannot understand others' false beliefs, remember their own false beliefs, or distinguish appearance from reality, except maybe with contextual support, but sometime between the fourth and fifth birthdays, most kids can do this on their own fairly easily. It would be best to describe each test in its own critique, rather than go over all of them here, but there are a few basic formats:

1. Appearance versus reality: a sponge made up to look like a rock. Older kids can tell it looks like a rock but is a sponge, while for younger kids it is what it looks like, either a sponge or a rock depending on their mood.
2. Deception: Hide chocolate in one of two boxes, so that the child knows which but another person doesn't. The child passes the test if he or she lies to the other person, telling them to look in the empty box, so they can keep the chocolate (or coin, in another version) themselves.
3. False beliefs: This is the big one, the test that all people who discuss theory of mind need to be familiar with. The child knows an object has been moved from one place to another, or that one object has been substituted for another. Someone who doesn't know about the move or substitution will naturally think differently, and will be surprised to be wrong. Children pass the test when they know the other person will be fooled, and in some cases, that they themselves were fooled, too, before they were let in on the joke. Children too young to pass the test think the other person won't be fooled, and don't even remember that they were fooled themselves.

According to Astington, Francesca Happé found that there is a strong correlation between language ability and ability to pass the theory of mind tests (I haven't actually read this paper myself, so I'm not sure how accurately it's reported and what it's taken to mean). Also, research by Perner and colleagues shows that regular kids do better at theory of mind tasks if they come from larger families and presumably have more intense social interaction and therefore more practice at social cognition growing up. Astington's own research indicates that with ordinary kids, kids with better language ability and memory capacity (which generally go together) do better on false belief tests. But then she goes on to suggest (repeating the findings of Baron-Cohen and others) that autistic people lack theory of mind, which leads to the language and communication deficits typical of autism, instead of considering the possibility that autistic people have language and communication deficits, which leads to them doing less well on theory of mind tasks. And she'd already indicated that most autistic people do pass these tests at some point to some degree, so they can't lack theory of mind wholesale. Sigh. Do I have any hair left?

Other than her tendency to believe everything other researchers say about autistic people, Astington has written a good basic summary of theory of mind research, and it's worth taking a look (but a cautious one, and remember to have those drugs handy).

Source

Janet Wilde Astington, 1993. The Child's Discovery of the Mind. The Developing Child Series. Harvard University Press.