34 | NewScientist | 8 June 2013If you think that explanation sounds
complicated, you’re not alone. “The key
problem is explaining why you would bother
acquiring the same concept twice,” says
Rebecca Saxe, a cognitive scientist at
Massachusetts Institute of Technology.
Yet there are other mental skills that
develop twice. Take number theory. Long
before they can count, infants have an
ability to gauge rough quantities; they
can distinguish, for instance, between a
general sense of “threeness” and “fourness”.
Eventually, though, they do learn to count
and multiply and so on, although the innate
system still hums beneath the surface. Our
decision-making ability, too, may develop
twice. We seem to have an automatic and
intuitive system for making gut decisions,
and a second system that is slower and
more explicit.Double-think
So perhaps we also have a dual system for
thinking about thoughts, says Ian Apperly,
a cognitive scientist at the University of
Birmingham, UK. “There might be two kinds
of processes, on the one hand for speed
and efficiency, and on the other hand for
flexibility,” he argues (Psychological Review,
vol 116, p 953).
Apperly has found evidence that we still
possess the fast implicit system as adults.
People were asked to study pictures showing
a man looking at dots on a wall; sometimes
the man could see all the dots, sometimes not.
When asked how many dots there were,
volunteers were slower and less accurate if
the man could see fewer dots than they could.
Even when trying not to take the man’s
perspective into account, they couldn’t help
but do so, says Apperly. “That’s a strong
indication of an automatic process,” he says –
in other words, an implicit system working at
an unconscious level.
If this theory is true, it suggests we should
pay attention to our gut feelings about
people’s state of mind, says Apperly. Imagine
surprising an intruder in your home. The
implicit system might help you make fast
decisions about what they see and know,
while the explicit system could help you to
make more calculated judgments about their
motives. “Which system is better depends
on whether you have time to make the more
sophisticated judgement,” says Apperly.
The idea that we have a two-tier theory
of mind is gaining ground. Further support
comes from a study of people with autism,
a group known to have difficulty with social
skills, who are often said to lack theory of
mind. In fact, tests on a group of high-
functioning people with Asperger’s syndrome,
a form of autism, showed they had the explicit
system, yet they failed at non-verbal tests ofthe kind that reveal implicit theory of mind in
babies (Science, vol 325, p 883). So people with
autism can learn explicit mentalising skills,
even without the implicit system, although
the process remains “a little bit cumbersome”
says Uta Frith, a cognitive scientist at
University College London, who led the work.
The finding suggests that the capacity to
understand others should not be so easily
written off in those with autism. “They can
handle it when they have time to think about
it,” says Frith.
If theory of mind is not an all-or-nothing
quality, does that help explain why some of
us seem to be better than others at putting
ourselves into other people’s shoes? “Clearly
people vary,” points out Apperly. “If you think
of all your colleagues and friends, some are
socially more or less capable.”
Unfortunately, that is not reflected in the
Sally-Anne test, the mainstay of theory of
mind research for the past four decades.
Nearly everyone over the age of 5 can pass
it standing on their head.
To get the measure of the variation in
people’s abilities, different approaches are
needed. One is called the director task; based
on a similar idea to Apperly’s dot pictures, this
involves people moving objects around on a
grid while taking into account the viewpoint
of an observer. This test reveals how children
and adolescents improve progressively as they
mature, only reaching a plateau in their 20s.
How does that timing square with the fact
that the implicit system – which the director
test hinges on – is supposed to emerge in earlyinfancy? Sarah-Jayne Blakemore, a cognitive
neuroscientist at University College London
who works with Apperly, has an answer. What
improves, she reckons, is not theory of mind
per se but how we apply it in social situations
using cognitive skills such as planning,
attention and problem-solving, which keep
developing during adolescence. “It’s the
way we use that information when we make
decisions,” she says.
So teenagers can blame their reputation
for being self-centred on the fact they are still
developing their theory of mind. The good
news for parents is that most adolescents
will learn how to put themselves in others’
shoes eventually. “You improve your skills
by experiencing social scenarios,” says Frith.
It is also possible to test people’s explicit
mentalising abilities by asking them
convoluted “who-thought-what-about-whom”
questions. After all, we can do better than
realising that our friend mistakenly thinks her” We may have a two-tier
theory of mind – one
process for speed, the
other for flexibility”Most 3-year-olds think that Sally will look in the box.
Around the age of 4 or 5, children understand that
although they know the ball has been moved, Sally
does notThe standard way to see if young children have
acquired this ability is the Sally-Anne testDepending on how the test is presented
20-50% of adults would incorrectly move
the topmost ballWe need more di cult tests for
adults Some objects are screened from
Sally’s view, so you must take this into
account when following her instructionsMove the small ball
down one levelSally puts her ball in her basket,
covers it and walks awayAnne takes the ball from the basket
and puts it in her boxWhen Sally comes back, where will
she look for her ball?Mind reading
Understanding that other people can have knowledge
or beliefs that di er from our own is a crucial part of
our social intelligence130608_F_Mindreading.indd 34 31/5/13 14:24:05