Psychologists in general, and developmental psychologists in particular,
suffer from a special dilemma. All human beings have intuitive ideas about
minds, including their children’s minds. Scientific psychology is measured
against this intuitive psychology – and often seems either obvious or absurd
as result. Both sciences, furthermore, are afflicted by anxiety about being
a “soft” rather than “hard” science. In reaction to these dilemmas and
anxieties, developmental psychologists often divorce their science from
their lives. We say that studying children is just a means to an end – that
we’re more interested in answering the deepest questions in cognitive
science than in children themselves.
But all great scientists start out by being captivated by the subjects they
study, and all the great developmentalists start from a sense of wonder at
children’s everyday lives. Cognitive development began, after all, with Jean
and Valentine Piaget’s amazingly rich observations of their own children.
Parents and children live, feel, think and act together in rich, complex and
enormously significant ways. How does the science of developmental
psychology interact with the everyday experience of caring for children?
Charles Fernyhough’s The Baby in the Mirror and Vasudevi Reddy’s How Infants
Know Minds raise this question in different ways. Fernyhough’s book is a
memoir of his daughter’s early childhood, interspersed with information
about developmental psychology. Reddy’s book is essentially academic. She
explicitly argues that we should reject the idea of an objective
developmental science in favour of a more engaged “second-person” approach.
Both books provide exceptionally sensitive, careful and thoughtful
descriptions of the everyday lives of babies, particularly the authors’ own
babies. Reddy’s book is full of eloquent and informative descriptions of the
playful way that even young infants tease, act coy, and generally muck about
with their parents. Fernyhough is primarily a novelist, and his book is an
elegantly written, warm, thoughtful, novelistic account of his first three
years with his daughter Athena.
Childhood is central to many memoirs and novels, but good descriptions of very
early childhood, good stories about babies, are surprisingly rare. Perhaps
it is because becoming a parent is so emotionally overwhelming that it
undermines the detachment that is necessary for either literature or
science. Both Reddy and, especially, Fernyhough do a lovely job of conveying
what life with a baby is like. Neither book, however, is very effective at
conveying what the science of cognitive development is like.
Cognitive development, like the rest of cognitive science, relies on a general
theoretical framework. That framework assumes that the mind performs
computations and the brain implements them. Developmental cognitive science
asks, and answers, some of the most important questions about those
computations. Where do they come from? Are they innate, or the result of
simple associations? Or are they derived from experience by some kind of
rational inference? How much do they depend on social interaction,
maturation, perception and action? Cognitive development has actually been
the most cognitive branch of cognitive psychology. While most cognitive
psychology asks what representations are like, or how they are stored,
manipulated and used, cognitive development asks where those representations
come from and how they allow us to know about the real world.
Within that framework there is room for often ferocious theoretical debate.
Moreover, the framework itself is profoundly unintuitive – it doesn’t look
much like our intuitive psychology, and certainly doesn’t look like our
intuitive psychology of babies. Babies just don’t look like creatures with
innate modules, or inductive inference engines, or associationist networks,
or intuitive theories, and that’s not how we treat them in our everyday
life. But that general framework has been impressively productive. It has
generated startling empirical results and raised intriguing new theoretical
possibilities.
Fernyhough is a part-time lecturer in psychology as well as a novelist, and
the scientific information in his book is accurate and clear as an
introductory course in psychology would be – here’s some Piaget and Vygotsky
and Chomsky, a little neuroscience, some recent experiments. But the links
between the science and Fernyhough’s everyday experience of his daughter are
frustratingly unclear. To take just one example, he makes much of the fact
that, according to Piaget, infants are unable to either remember specific
past events or represent absent objects. Then a few pages later he outlines,
approvingly, Elizabeth Spelke’s ideas about core knowledge. Spelke thinks
that infants have essentially adult representations of objects and people.
The trouble is that if Spelke is right, then Piaget is wrong, and vice
versa. Fernyhough writes as if baby Athena is like Piaget’s baby, a creature
with a mind of sand that leaves no trace of its experience. But according to
Spelke, Athena knows just as much about absent objects as Fernyhough does.
Even the brightest and most sensitive dad can’t decide between these two
possibilities based on his everyday experience, and Fernyhough doesn’t
really try to. It’s as if he free-associates from his everyday experience to
his developmental reading, rather than trying to think through just what
that the scientific possibilities are, and what they would mean for
understanding Athena’s mind.
Reddy, on the other hand, and writers like her, argue that the entire
theoretical framework of cognitive development is mistaken. For some
psychologists and philosophers it seems prima facie impossible to capture
the dynamic, fine-grained and beautifully complex interactions between
people in general, and parents and babies in particular, in these abstract
terms.
It is healthy to question the broad presuppositions of a discipline from time
to time. But, in fact, understanding social interaction has been at the
centre of many of the most exciting recent studies in cognitive development.
The topic of Reddy’s title, “How infants know minds”, is just the question
that they’ve been trying to answer. The question is particularly important
because it is also the answer to the question of how all of us know minds,
since we are, after all, only infants who have been around a while.
Understanding other minds, as Reddy herself agrees, is at the core of
everyday social life. This becomes vivid when you consider the profound
social difficulties of people with autism, who have trouble understanding
minds. But cognitive scientists have explored this question using rigorous,
careful, objective experiments, and explained them in terms of
representations and rules – just the method and the theory that Reddy
rejects. Reading Reddy’s wonderful descriptions of baby coyness and teasing,
you find yourself just itching to get her to do some experiments and some
theorizing, and find out what is really going on.
In the late 1980s, cognitive developmentalists began to ask a question
empirically that up till then had been the province of philosophy – how do
people come to understand the minds of others, and their own minds too? The
unexpected answers to that question shaped a whole field, “theory of mind”,
which has more recently been adopted by neuroscience, adult cognitive and
social psychology. Although the field has come to be identified with a
single experimental task, the ubiquitous “false belief” problem,
developmentalists actually used a wide range of observational and
experimental methods to answer that question. For example, Karen Bartsch and
Henry Wellman did beautiful and informative analyses of children’s spontaneous
everyday talk about the mind. This kind of everyday conversation turned out
to confirm the experimental results while the experimental results helped to
explain how children talked.
From the start, psychologists studying “theory of mind” suspected that very
young infants must know something about other people’s minds. As Reddy
points out, this was because, in the 1970s, Colwyn Trevarthen, Jerome Bruner
and others described surprisingly sophisticated social interactions between
babies and parents, interactions that seemed to require knowledge about
other minds. This work reached a kind of apex in Andrew Meltzoff’s
experimental discovery that newborn infants imitate adult facial
expressions. But the observational and experimental techniques used to study
theory of mind in young children wouldn’t work with babies. On the other
hand, “looking-time” techniques had been used very successfully to study
infant perception. These are studies that measure the amount of time that
babies look at some scenes rather than others – babies look longer at
unexpected events. But these studies are notoriously difficult to interpret,
and the results often seem unrelated to the kind of knowledge we see in
babies’ actions or interactions.
Several pioneering researchers, however, have started to find out what babies
do know about the mind, when they know it, and, most significantly, how they
figure it out. For example, Amanda Woodward showed seven-month-old babies a
hand reaching for one of two objects, say, a ball or a teddy bear. Then she
switched the location of the two objects, and the babies saw the hand reach
for the new object at the old location, or the old object at the new
location. If babies think the hand belongs to a person who wants the object,
then they should be surprised if they see the hand reach for a new object,
and they should look longer at that event. That’s just what they did. When
children saw a stick move in the same way, they didn’t distinguish the two
events. They seemed to think that the person was intentionally reaching for
an object, but the stick just touched it.
Next, Woodward, along with Jessica Sommerville and Amy Needham, related this
result to babies’ everyday actions. Some young babies can reach for objects
and pick them up and some can’t. Woodward and Sommerville discovered that
babies who could reach for objects themselves were more likely to think that
the other person was reaching for an object. These observations provided
converging evidence that the original finding was genuine, and not just a
reaction to superficial perceptual features of the events. This study also
suggested that babies might actually learn about the intentions of others
through their experience of their own actions. In the final, most elegant,
and most significant study, they gave three-month-old babies “sticky
mittens”, Velcro-covered gloves that allowed the babies to manipulate Velcro
objects. Three-month-olds usually can’t reach objects but the sticky mittens
magically gave them this ability. Even three-month-old babies who got this
experience made the right inferences about the other person in the
looking-time experiments.
As these elegant studies show, developmental psychologists can integrate
observational and experimental data to make real discoveries about what
young babies know about other people. Those discoveries, in turn, can answer
some of the questions that a sensitive father like Fernyhough might ask.
Yes, seven-month-old Athena really could understand what her dad wanted.
They also answer some of the questions that Reddy thinks cannot be answered
by developmental science. Yes, infants do know at least something about
minds. And they find out about minds by making inferences from their own
experiences and their experience of others.
The “sticky mittens” study, especially, is representative of the most exciting
breakthrough in cognitive development in a long time – one that neither
Reddy nor Fernyhough seems to appreciate. For many years developmental
psychologists primarily studied what children know when. In the past ten
years we have also started to understand how they know it. Empirical studies
have shown that even the youngest babies learn systematically from their
experiences, and computational theories of learning have started to provide
rigorous accounts of how that learning is possible.
That kind of science makes our everyday experience of children even richer.
Perhaps it will also give us psychologists the confidence to proudly embrace
our study of the soft, warm and occasionally fuzzy creatures we call people,
and the even softer, warmer and fuzzier ones we call babies.
Charles Fernyhough
The Baby in the Mirror
A child’s world from birth to three
273pp. Granta. Paperback, £12.99.
978 1 84708007 3
Vasudevi Reddy
How Infants Know Minds
273pp. Harvard University Press. £22.95 (US $35).
978 0 674 02666 7
Alison Gopnik is Professor of Psychology at the University of
California at Berkeley.
NB especially V Reddy
tanya hart, hong kong, hong kong