Social: Why Our Brains Are Wired to Connect (21 page)

In order to test her account,
Heyes designed a clever
countermirroring
procedure
during which the performance of an action was associated with a
different
visual action.
If I am instructed to move my foot whenever I see a hand moving, a true motor resonance mechanism should be insensitive to this behavior.
Yet Heyes found that when individuals learn to respond with a behavior different from the one seen, the mirror system is activated just as it is during direct imitation.
This outcome suggests that while the mirror system frequently responds to the sight and performance of the same action, this response is not intrinsic to its functioning; it can learn to respond to the sight of one action and the performance of a different action.
If mirror neurons link my foot movement to your hand movement, it’s hard to see how this response constitutes motor resonance or promotes mindreading.

Another study examined how the mirror system responds
when individuals perform a complementary behavior instead of imitating an observed behavior.
Imagine that there are two objects on
a table: a small object that you pick up by pinching your thumb and first finger together (for example, a sugar cube) and a larger cylinder that you pick up using a cup-like grip (for example, a can of soup).
In some trials, individuals were told to imitate the motion and grip used by the other person shown in a video.
For other trials, individuals were told instead to complement the other person by preparing to grab the other object using the other grip.
Here, if your partner started to pinch to pick up the sugar cube, you would make a cup-like grip to pick up the can of soup.
The mirror system was more active during the complementary action trials than it was during the imitative action trials.
There is no reason why a motor resonance mechanism would be more active as we engage in a behavior different from the one seen.
Along with the previous study, this suggests that the mirror system is not designed solely for the purpose of matching our internal state to that of another person.

What Are Your Intentions?

There have been dozens of mirror system studies of mindreading inspired by Simulation theory and dozens of mentalizing studies inspired by the Theory of Mind accounts.
Both research programs have focused on how one person understands the mental states of another person.
One might think that whatever one’s theory going in, everyone would be looking at the same brain, so the results of these studies would have to converge and tell one story.
Although the studies were motivated by different theories, they still should have ended up in the same place because the brain does not care about our theories.
It should just show what’s true.

Yet despite the numerous fMRI studies from both the mirror and the mentalizing camps, their results almost never come together.
From looking at the brain scans of different studies, you wouldn’t guess that all of these researchers have been studying mindreading.
Not only have the two camps consistently produced findings that
do not overlap anatomically, but the regions of the brain observed by each group tend to be inversely correlated with one another.
If we look at brains at rest
, the more activity people produce in the mentalizing system, the less they produce in the mirror system.
These were not just opposing theoretical camps.
They were studying what appeared to be opposing neural systems.
Yet both systems are supposedly end-to-end solutions for mindreading in which the other system plays no role.

There are two reasons why each camp saw only their preferred regions active when they were ostensibly studying the same thing.
First, the two camps study mindreading in very different ways.
The mentalizing camp tends to use verbal materials and cartoonish drawings.
In other words, the materials are pretty abstract.
If the mirror system is activated by seeing real action, it is understandable that mentalizing studies not showing real action would fail to activate mirror regions and overlook their contributions to mind-reading.
On the flip side, a major strength of mentalizing studies is that they manipulate whether a person is trying to understand the mind of another person.
After reading a paragraph that implies an individual’s mental state, participants in mentalizing studies are often asked questions about the protagonist’s beliefs, motives, and personality—questions that can be answered correctly only if the proper mentalistic inferences have been drawn.
Mirror system studies never pose these mental state questions to participants, perhaps because when you see only a disembodied arm, questions about beliefs and personality do not make much sense.
Thus, mirror system studies minimize the involvement of the mentalizing system.

The second big issue, the Clintonian verbal hairsplitting issue, concerns the meaning of words like
goal
and
intention
.
Let’s say you see a friend drinking a glass of single malt scotch at 8 a.m.
You ask him why.
If he answers, “In order to have a drink,”
strictly speaking, he is answering your question, providing you with a goal (“to have a drink”).
But his answer is entirely unsatisfying.
It is obvious
that he is taking a drink because he wants to take a drink.
What you really want to know is what special motivation led him to the unusual goal of wanting to have an alcoholic drink at this hour of the morning.
The responses “to have a drink” and “to drown my sorrows because I lost my job” are technically both answers to the question, but they highlight how the word
goal
can have different meanings.

In the 1980s, Robin Vallacher and Daniel Wegner
systematically investigated these distinctions
.
They conducted a series of studies highlighting how we can understand the same action in different but equally accurate ways.
I can describe my current behavior at my computer keyboard as “moving all of my fingers slightly up and down,” as “typing,” as “writing a book,” or even as “trying to share what I have learned with others.”
Given the order in which I’ve presented these options, we can see that they form a hierarchy, with the first answers giving lower-level descriptions of specific motor behaviors and the latter ones describing higher-level long-term goals that have greater meaning.
No one comes to the end of their life and says, “I wish I had moved my fingers up and down more,” but we can imagine someone saying, “I wish I had taken the time to share more of what I learned.”
We live in a world of meaningful actions that can be described at both high and low levels, but we typically focus on one level at a time, depending on what we are interested in.
New typists focus on which fingers they are moving to find particular letters, while experienced typists
are more likely to focus on the thoughts they are trying to convey
.

Among the biggest differences between mirror neuron researchers and mentalizing researchers are the kinds of goals they are interested in explaining.
Mirror neuron folks are focused on how we understand the lower-level motor intentions of others (“He is flicking the light switch because he wants the light to come on”), whereas the mentalizing folks are more interested in higher-level intentions (“He is turning on the light because he wants to study for
an exam”).
The intentions of the other person are described in both cases, but I think it’s fair to say that in everyday life we are usually interested more in the second kind of goal than in the first.

The motor resonance account is well positioned to explain how we understand low-level motor intentions.
I see you flicking the light switch, which activates the “light switch–flicking” mirror neurons in me.
But these same neurons are poorly positioned to
explain another person’s high-level reasons for wanting the light on
.
There are countless reasons why I might want the light on, and many of these are belief based (“I heard a noise downstairs in the middle of the night so I am turning on a light to see if anyone is there” or “I woke up with a great idea for a story so I want to turn on the light in order to write it down”).
We don’t turn on the light differently in each of these instances, so there is no way for motor resonance to clue us in to which of these higher-level intentions is present.
The mentalizing system of the brain is ultimately needed to figure out the higher-level intentions of others.
The question is, what role, if any, does the mirror system play in allowing us to understand the higher-level motives of those around us?

How, What, and Why

When we look at another person’s behavior, there are three questions that we might be interested in answering.
These three questions correspond to different levels of analysis in the Vallacher and Wegner studies.
The first, obvious question is
what
someone is doing, which we answer in the most generic action language possible: “She’s crossing the street,” “He is typing,” “The cat is eating my leftovers.”
We interpret actions this way so regularly that we don’t even realize we are doing it unless someone does something really out of the ordinary (“Is he scaling the side of a building?”).
Depending on our goals, we might then ask ourselves one of two follow-up questions.
If we have any further interest in the person,
we most likely want to know
why
the person is doing
what
they are doing: “She is crossing the street
in order to
get to work,” or “He is typing
in order to
finish his final paper.”
Sometimes, though, our interest is less in the person and is instead focused on the behavior itself.
We may want to learn
how to perform
the same behavior, such as when a student taking guitar lessons watches his instructor in order to figure out
how
the teacher is doing
what
he is doing.

Bob Spunt and I have conducted a series of studies to examine how the mirror and mentalizing systems contribute to our
figuring out the how, what, and why of other people’s behavior
.
Do different systems in the brain handle each of these questions?
We intuitively expect different parts of the brain to handle seeing and hearing because the experiences of seeing and hearing are so fundamentally different from one another.
I’m not so sure that looking at the same behavior in others and asking
how, what
, and
why
seem different enough to involve distinct regions of the brain.
But this is why we actually run the studies.

In undertaking the studies, our thinking was that if you see a behavior (for example, a woman recycling a bottle) and ask the question “Why is she doing that?,” your answer will likely involve a high-level meaningful answer that requires the mentalizing system of the brain (for example, “She is a conscientious person,” “She wants to protect the environment,” or “She wants to impress a guy who she knows recycles”).
However, if you simply want to imitate the behavior and ask “How is she doing that?,” this response would involve low-level answers that do not require the mentalizing system but would instead rely on the mirror system of the brain that represents motor movements (for example, “She puts the glass and plastics in the blue bin”).
This is exactly what we have seen in multiple
why-how
studies.
Whether participants are watching everyday actions or watching someone experience strong emotions, asking
why
recruits or draws on the mentalizing system, while asking
how
recruits the mirror system.

We also wanted to find out what the brain was doing when a
person answers the
what
question.
With the
why
and
how
questions, we just instructed people to answer those questions.
But given that people answer the
what
question in everyday life without stopping to think about it most of the time, we thought that prompting people to stop and think about it might not actually capture the natural process.
The way we worked around this was by manipulating whether or not participants needed to spontaneously answer the
what
question before they could answer the
why
or
how
question they were actually given.
Sometimes we showed participants a video of a person performing a task (for example, a woman highlighting passages in a textbook), and other times we replaced the video with a description of the action (for example, “She is highlighting passages in a textbook”).
Whether answering the
why
or
how
question, the first thing the participant needs to do when the action is shown visually is to figure out
what
is happening.
In contrast, when the action is described in words the
what
is already characterized in the description of the action—it is literally the answer to the
what
question.
By comparing the visual and verbal presentations of the actions, we can identify brain regions that support the implicit
what
decoding process.

When we looked at the brain to see what regions were more active during visual than during verbal presentations, we saw two things.
First, we saw lots of activity in the back of the brain in the visual cortex.
This was expected, as video clips contain far more visual information than text.
The other thing we saw was increased activity in the mirror system.
Whether the explicit task involved asking
why
or
how
,
there was increased mirror system activity
when the action was presented visually.
In fact,
this was still the case when participants were distracted
by being asked to recite a seven-digit number while performing the
why-how
task.
Distracting people in this way is a common technique for identifying processes that are so automatic that they still take place despite participants being distracted.
The fact that even with this distraction, the mirror system responded to the visual action suggests that its decoding
of
what
is happening is pretty automatic.
In contrast, the mentalizing system was significantly less active when people were under cognitive load, suggesting that this system does not do well when people are distracted.