Social: Why Our Brains Are Wired to Connect (25 page)

Although early findings have been provocative, it isn’t clear how strongly they support the broken mirror hypothesis.
For instance, an early study focused on
mu suppression
, a biomarker of mirror system activity measured with an electroencephalogram (EEG).
Normally developed individuals produced mu suppression both when observing and performing hand actions.
However,
individuals with autism produced mu suppression only when they performed hand actions themselves
, not when they observed them.
Oddly, though, the researchers did not report whether there was a significant difference between the two groups.
This might seem like a minor oversight, but without this analysis,
the conclusion that mirror system activity differs in the autistic sample is unwarranted
.

Two early fMRI studies also suggested atypical mirror systems in individuals with autism, but not in ways that clearly map onto the symptomology of autism.
The mirror system has a frontal component (the posterior inferior frontal gyrus and the premotor cortex) and a parietal component (the anterior intraparietal sulcus and the rostral inferior parietal lobule) further back in the brain (see
Figure 6.1
).
One fMRI study found that individuals with autism
produced decreased activity in the frontal component of the mirror system but increased activity in the parietal component when
imitating facial expressions.
The other study found that this group produced decreased activity
in the parietal component of the mirror system when imitating hand actions but increased activity in the frontal component.
In both of these studies it was clear that the brain was doing something different in individuals with and without autism during imitation, but each was a mix of increases and decreases, and these differences were in the opposite direction in the two studies.
Perhaps the bigger issue was that in both studies, the autistic individuals were able to imitate as well as the nonautistic participants, so it was unclear whether the neural effects seen were contributing to any real-world behavioral problems in autism.

Since these early studies, a number of other studies have contradicted them.
Multiple studies have shown roughly equal levels of mu suppression
in autistic and healthy samples.
Similarly,
a number of fMRI studies have shown equivalent or enhanced mirror system
activity in individuals with autism.
So why are these results so messy and inconsistent?
If there are imitation deficits in autism, there ought to be mirror system deficits as well, right?
Maybe not.

Victoria Southgate and Antonia de Hamilton provided a compelling explanation of why
imitation performance cannot be so easily equated
with the mirror system in autism.
They pointed out that in real life and even in the lab, successful imitating is about more than, well, imitating.
To be successful, you need to know what to imitate and when.
If an experimenter says “Do this” and then picks up a pen from the table, what exactly are you meant to do?
Is the key to pick up the pen?
To pick it up with the same hand as the experimenter?
To use the same grip or approach it with the same movement through space?
The experimenter might code the failure to do any one of these things as poor imitation.
Imitating well requires knowing what should be imitated and then being able to implement that precisely.
Knowing what to imitate, particularly in lab studies, is about understanding what someone wants you to do.
This is a mentalizing task, which depends on Theory of Mind, something we already know is impaired in autism.

Figure 7.3 The Hand Gestures Used During Automatic Imitation.
The thumb and fingers making a U shape (left), a wide-open hand (middle), and a closed fist (right).

One way to solve this problem is to eliminate the mentalizing component from the imitation process.
Scientists have done this by examining involuntary imitation.
Sometimes we imitate others without intending to or even when we know we shouldn’t.
Imagine the following.
You see a hand making a U with the thumb and fingers (see
Figure 7.3
).
When you see the thumb and fingers start to come together into a closed fist, you are meant to do the same with your hand.
Next, you see the U-shaped hand again, but this time it opens up as wide as possible, and you are supposed to do that as well.
So far, so good.
This is a simple imitation task.
Here’s where it gets interesting.
Sometimes you are asked to imitate what you see (that is, to make a fist when you see the target making a fist), and other times you are asked to do the opposite (that is, to open your hand wide when you see the target making a fist).

It is hard to do the opposite of what you see because it’s natural to imitate.
As a result,
nonautistic individuals take longer to perform the incompatible movement
(doing the opposite of what they see) than to perform the compatible movement (doing what they see), and this is an indicator of the strength of our automatic imitation tendencies.
In the first study to employ the automatic imitation task in individuals with autism, this group not only produced the automatic imitation effect, but the effect was nearly 50 percent stronger in the group with autism compared with the nonautistic participants.
Other studies have produced similar effects.
Another group using a different automatic imitation paradigm
also showed evidence of hyper-imitation associated with autism.
Once the mentalizing component is stripped away, rather than failing to imitate, individuals with autism may actually be hyper-imitators.

The broken mirror hypothesis is still relatively new so it’s unclear how history will judge it.
For the time being, the weight of evidence is against it, with some of the most recent data suggesting that autistic individuals are hyper-imitators who do not know when or exactly what to imitate when explicitly asked to do so.
If autistics’ imitation difficulty really boils down to having trouble understanding what is to be imitated, this would return us to the well-established finding of deficient Theory of Mind in autism.
But we still need to find out why these individuals have this deficit, given that the social impairments exist prior to the development of Theory of Mind abilities, and we know from the study of deaf children that Theory of Mind can develop poorly because of one’s experiences alone.
Fortunately, we have one more bowl of porridge to try, and hopefully it will be just right.

Intense World Hypothesis

Remember my “bad trip” at Rutgers?
There’s a second half to that story, the internal half.
I described my behavior earlier—what others around me could see.
Basically, I kept my distance from everyone and came across as uninterested in my social environment.
Obviously I had not become autistic, even for a little while.
But there is a lesson to be learned from my experience.
I call it the
Head & Shoulders effect.

You might remember the Head & Shoulders advertising campaign from the 1980s.
In these ads there were always two people, one of whom noticed the other’s dandruff.
The person who saw the other’s flakes pointed it out, then said something along the lines of, “Hey, try my Head and Shoulders,” and then passed the conveniently on-hand shampoo bottle to the flake-afflicted friend.
The receiver always responded by saying, “But you don’t have dandruff!”
to which the first individual flashed a winning smile and said, “Exactly.”
The implication was that because he used Head & Shoulders, no one would ever know he had dandruff.
The psychological parallel is that how things look on the outside are often the opposite of what is going on inside the person.
We tend to assume that outsides and insides match
, but because people react to their circumstances and try to compensate, they often do not.

On that fateful day at Rutgers, people could not have guessed my internal experience from my outward behavior.
My behavior and experience were intimately connected, but in a nonobvious way.
My outward behavior was antisocial, but this is not because I was uninterested in the social world.
Rather, I was
overwhelmed
by the social world.
To put a fine point on it, I was overwhelmed by everything, but the social world was the most overwhelming part of all the overwhelming experiences that day.

The drug that I had ingested heightened all my senses.
Usually that is pretty cool, but as the guys from
This Is Spinal Tap
might say, my senses were turned up to eleven.
It was all just too much for me to process.
All the perceptions we have that are usually in the background were suddenly in the foreground and much too intense.
An unmoving frozen environment would have been distressing enough, but I was surrounded by people making noise and full of facial reactions, gestures, and other sudden movements.
All of it was very intense to me, surprisingly unpredictable, and frankly, terrifying.
I was antisocial that day not because of a dislike of those people but rather because of the simple fact of them being people, which made them too much for me to handle that day.

What if children with autism, rather than being insensitive to the social world, are actually too sensitive to the social world?
What if the distressing intensity of early social interactions leads these one- and two-year-olds to prefer isolation over social contact?
If overly intense experiences promote social isolation, these children might go on to miss out on countless interactions that would train
their brains to become social experts over the next decade.
Perhaps those with autism are simply cutting class because it is too painful to stay in class.
You have probably had the experience of covering your ears when a movie theater demos its sound system before a movie (my wife and son both do this, every time).
If your life were always like that, wouldn’t you find a quieter place?

I think it is possible that autistic adults are less socioemotionally sensitive, at least in some ways.
Even if this is true, the question is whether they were always insensitive to the social world because of their genetic dispositions or whether the lack of social engagement is acquired as a result of the autistic individual’s rational responses to a childhood hypersensitivity to the social world.
This is the essence of
the
intense world hypothesis
of autism
.
The distress of early life leads these children to turn away from the social world, causing them to miss key social inputs that ordinarily would help the mentalizing system mature.

The intense world hypothesis is relatively new and counterintuitive.
Is there any evidence for it?
Within the autistic community, people have certainly reported feeling this way.
Jay Johnson is a blogger who has autism and writes about this experience in the context of explaining why he does not make eye contact:

People are loud confusing creatures… . And they expect me to add eye contact?
I actually don’t know how it feels for you, but for me, looking into another person’s eyes and having them look back into mine
feels like I am touching a hot stove.
I am being burned.
It’s like an extra jolt of overwhelming input.

Beyond this sort of anecdotal experience, there is also a fair amount of empirical support, even though few researchers are actually looking to promote this counterintuitive theory.
Given the tendency we all have for confirmation bias (searching only for what we hope to see), perhaps this is a good sign for the hypothesis.

Early on, it was believed that individuals with autism might be
less emotionally sensitive and that part of the reason for this might be diminished amygdala sensitivity.
As discussed earlier in the chapter, the amygdala is a small structure that responds to and codes for the emotional intensity of events in our environment.
In humans, the amygdala seems particularly responsive
to social inputs such as other people’s emotional expressions.
And while the amygdala does respond to intense positive and negative cues
in the environment, there is reason to think it is more central to negative emotional experiences, like fear and anxiety.
Even subliminally presented fearful faces
that an individual never reports seeing reliably turn on the amygdala.

The best initial evidence for an amygdala-autism link
came from studies comparing the neural responses of adults with and without autism to faces expressing emotions like fear or anger.
The most consistent finding from these studies was that autistic individuals produced a weaker amygdala response to these threatening social cues.
When these findings were combined with the fact that amygdala damage in nonhuman primates
can produce some autistic-like characteristics, it seemed to follow that autistic individuals might be somewhat insensitive to the social world because their amygdala wasn’t tuning in and directing their attention to the social world.