Amy Chua, Neurotiger

“Why is Amy Chua so upsetting?”

Someone asked me this over the holidays while thrusting towards me me a copy of Amy Chua’s most recent defense of Tiger Mothering in the Wall Street Journal.

Now if we had been in my office, instead of at a cocktail party, I would have pulled the Shrink’s Delight.  The Shrink’s Delight is the passive reflection of a question back onto the patient.  I would have asked “well what comes to mind about that?” The person would have then gone on a tear about their own tiger mother, or the fact that they wish they had had a tiger mother, or that they feel they aren’t enough of a tiger mother.

However as I was in public, and wanted to avoid celebrity psychiatry while using public-record events to think about subjective neuroscience, I took the more prudent path of reading Chua’s article as my interlocutor watched breathlessly.

When I was done I broke out my crayons and drew the following diagram (Figure 1).

Figure 1: An attuned mother-infant relationship

“This is what an attuned realtionship looks like when a child is, oh, say six months old. And this,” I said, drawing Figure 2,

Figure 2: The Relationship Described in the WSJ

“is how Amy Chua is describing her relationship with her eighteen year old daughter. Maybe you are expecting her to be a six-month-old type attuned mommy, and you are getting a mismatch error that you find upsetting?”

The person looked at me like I was nuts. I had, it seems, just used a private language.

I apologized and backtracked into explaining the neuroself framework. In the neuroself framework, we use a single language to describe both neurons and psychology, thereby eliminating the need to figure out how the brain “makes” the mind. It’s a mashup of cybernetics and information theory that is generally related to neuroeconomics and the Bayesian brain, but built for clinical use.

In this framework, what brains and minds are evolved to do is create adaptive predictions, observe if they’ve come true, calculate the error – the mismatch between ideal and real (technically, this error is entropy, and is measured in voltage).   Thus when you feel sad, hungry, tired, anxious, confused, what is “really” happening is you are getting a mismatch error between your fantasies (or fears) and reality.  In the neuroself framework, this error is the self, which is why you have such a diversity of selves – hungry, sleepy, sad, and so forth – each of which in turn can be reconciled to one another by higher-level selves that determine if and when any given error (neuroself) will or will not gain access to the motor apparatus.

Once it has calculated with an error, the neuroself has two extreme choices, and a host of compromise choices between them.  The extremes allow us to think about the larger middle.  On one extreme it can manage the error internally – through a wide variety of techniques including avoidance, compartmentalization, denial, rationalization  – all of which involve changing its prediction. Once the prediction is in line with observation, the error goes away. Thus when you learn to “tune out” the sound in a restaurant, or the feeling of your clothes on your body, or your own scent you are bringing prediction into line with observation – typically through central nervous system accomodation of peripheral nervous system signals.

Alternately, and from an energy standpoint far more expensively, the brain can change its observations while leaving the prediction alone. This is done through motor behavior, and ultimately it is motor behavior that is the purpose of the nervous system (as motor behavior is required for reproduction; Priests try to change their predictions about sex, rather than having it, but do not reproduce).

Our motor behavior – when we eat, walk to the store, make a phone call, kiss someone, speak up in class, make a purchase, plant a tree – alters the outside world.  In time, the payoff is that our prediction – that we will have a full stomach, hear someone’s voice, feel their lips on ours, earn the admiration of the teacher, own a widget, have a tree in our yard – comes true: observation matches prediction, and error zeroes out.

“You’ve lost me completely,” said my interlocutor, “what does this have to do with why Amy Chua’s so upsetting?”

So I explained that this framework is ideal for modeling relationships, including mother-daughter ones.

“The relationship is this red circle you’ve drawn?” they asked.

To which I said yes.

Take any given error – say, that your ten year old child is not yet in college.  Some portion of your nervous system has generated this error – because, fundamentally, it has been wired up to predict that your child will be a college student.  You now have an error on your hands that is going to last for years until your child either is a college student or you relax your ambition for them.  This error is a neuroself, in our terminology.

If we zoom in on this neuroself, we see that it can be represented as a circle (see Figure 3).  Let’s say that this is your neuroself, and that you have a ten year old daughter, and that you want her to get into college.  But let’s say that you’re not entirely sure she’s really going to like college; she kind of wants to be a circus performer, and train in trapeze, and college would only slow her down.  You might then be perfectly torn – on the one hand, you want to make sure she grows up to be a good student, but on the other hand you think maybe you should stop expecting her to be one. You might then pursue a mixed strategy for resolving your error – encouraging her to both study and practice her sport until she decides which one to do.  We can represent this by dividing your neuroself in half, with a red line down the middle. The delta (triangle, standing for “error”) P on the left represents your effort to change your prediction (P for prediction) of what your daughter will become, while the delta O represents your effort to change your observation (O for observation), maybe by giving her gold stars when she gets a B+ or above in a class.

Figure 3: A neuroself.

The key point here, for now, is this red line. The closer it is to the left, the less invested you are in changing your predictions, and therefore the more invested you are in turning your kid into a great student. The closer it is to the right, the less invested you are into changing your observations, and the more open you are to finding out what she wants to do with her own life.  There’s no moral preference for solving an error one way or another – this is just a way of conceptualizing the difference between a laissez-faire parent who puts little pressure on their kids, and a tiger mom.

According to the WSJ article, Amy Chua has manifested two extreme forms of parenting in her life. As shown in Figure 4A, when her kid was younger she strictly enforced a vision of childhood (no sleepovers, etc), but now she is hardly involved at all, and is just sitting back to see what happens (Figure 4B). She’s gone from changing her error through observation to changing her error through prediction.

Figure 4: Two Kinds of Mom

As you can see, by moving this red line around, you can indicate what kind of strategy a neuroself is pursuing.

“That’s fun,” said my interlocutor – “I can use that to model why I just had a whole piece of chocolate cake and now am regretting it – first I changed my observation, now I am changing the prediction that made me eat it in the first place.”

“You mean before you ate it you predicted you should have a piece of cake, and now the you’ve eaten it, you’re predicting you shouldn’t.”

“Yes – except making predictions about faits accomplises is a bad strategy.”

“No comment.”

“But in Figure 1, what’s the red circle?” they asked.

So I explained that the red circle is just the red line. The resulting space is meant to represent, generally, public world – the stage on which all human relationships play out.  In the neuroself framework, that’s what the social world is: a place for the buying and selling of errors.  When a baker puts his cupcakes in his window, he’s putting an error into the world – he wants money, and his cupcakes are his way of (he hopes) observing it.  When you buy a cupcake, you are putting an error into the world – you want a cupcake, and your money is your way of observing it.  You have therefore just traded errors in such a way that they solved one another, and both of your neuroselves can now disappear.

This economic view of motor action is something we’ll develop in other posts, but for now the key concept is that a “relationship” is whatever is inside the red circle.  If you just try to solve all your problems on your own – by thinking about them – but never share them with anyone – through word or deed (words are motor products, remember) then you are “keeping them private” rather than “investing in the relationship.”

As such, these circles allow you to model how two (or more) people are managing a relationship, which brings us back to Tiger mothering.

In Figure 1, which I will reprint here, we can practice this way of framing things.

Figure 1: An attuned mother-infant relationship

Here the child is entirely inside the circle, which means that as in Figure 4B, it is pumping all of its error into a common space for the mother to solve – it wants to observe a change of what is distressing it. This is why when a baby is hungry, or tired, or cold, or scared, or lonely et cetera, it cries.  Crying is an error signal, and more specifically an attempt to alter its observations by getting the mother to perform motor behavior. The mother is then supposed to feed, swaddle, reassure, hug et cetera the child until the error goes away.  Obviously mothers are not meant to have their needs met by their infant, and so the mother has to alter her predictions (of how much sleep she’ll get, etc) rather than asking her baby to change. It’s six monts old – it can’t.

I then told my interlocutor my guess was that they were upset by Chua’s WSJ article because they didn’t like the model of the relationship that she was offering, and I suggested we diagram it.

Most of the action comes from this quote:

My husband and I are probably the most hands-off college parents we know. We never ask Sophia what she’s going to major in or what she does at night, and we accidentally forgot about parents’ weekend. When we got a few stressed text messages from her about finals, we told her to relax, do what she always does, and she’d be fine. And she was.

On several levesl she describes a lack of communication, interaction, and attunement.  She is purposely not tracking what is going on in her daughter’s life. In mainstream psychology this might be described as a lack of empathy. I’d say this is a lack of error. It’s not clear if that’s good or bad – only that it is. 

We see this in the report that the daughter reached out with a “few” stressed text messages (was one alone insufficient to elicit a response?) and that the mother responded with what reads as a pat reassurance rather than a phone call to talk through the error.  Morevoer, the assertion that her daughter was “fine” afterwards is an indication that the error – in Chua’s mind – has been zeroed out.  This raises questions of how the daughter managed this neat trick – was her mother’s brief response in and of itself reassuring, in that it communicated how unworried the mother was?

Then, in the most fascinating portion of the paragraph, and consistent with the idea that she does not want to risk learning of any errors in her daughter’s life, she “forgets” about parents’ weekend. The word “forgets” is of course a cognitive framing of what happened;  what is not known is what this forgetting meant to her daughter.  It is plausible that the forgetting generated errors in her daughter’s neuroself, which I am not modeling; the point here is that Amy Chua isn’t registering them.

In mainstream psychology all of this would be interpreted as some kind of a defensive effort on Chua’s part; by framing this in informational terms, we can see that Chua is managing to avoid registering any errors, and therefore avoiding the need to solve them. In normal psychological language, she doesn’t know how her daughter feels.

This is a more striking feature of the piece in general.  Chua appears not to have interviewed her daughter about how she feels, let alone devised a clever way to learn her daughter’s unbiased feelings.  One time honored way moms do this is by asking their kids’ friends, teachers, tutors, siblings, and so forth in order to get the low-down on how they are doing.  In my social science research on prison inmates (whose views of their incarceration I wanted, but who knew they were being monitored when talking to me, and thus were incentivized to give me prosocial responses) I learned a way to deal with this problem is to ask the respondent what other people think of their situation. Chua could ask to ask “do your friends want to trade moms with you?” or “would you recommend tiger mothering to a friend who’s mother wasn’t a tiger mother?”  However, consistent with the idea that Chua does not want to pick up any errors that her behavior is causing in her daughter’s neuroself, none of this is reported.

Finally, a striking feature of the print version of the article was the photograph of her daughter, which is lower down on the web version – the third image down on the left.  Her daughter is kneeling down and therefore not at her full adult height, and is wearing a tiger headband, with tiger ears on it, clearly a snarky reference to her mother’s book – though the history of the photograph, and who bought, photographed, and asked the photo to be shown in the WSJ are unknown.This photograph brings to mind the photographs that could have been, but were not, included in the article: of the daughter standing up, wearing her own outfit, engaged in her own interests, and as her own person. The overall message is that the daughter is being presented through her mother’s eyes rather than “in her own voice,” which is merely further confirmation that the concern here is not with how the daughter herself has experienced all of this.

With this as our data, we can now return to Figure 2. I modeled the mother-daughter relationship this way because neither of them are pumping errors (as per the article) into the relationship – they are, as per the report, barely communicating.

Figure 2: Hands Off

In the neuroself framework, the “economy of errors” that drives most relationships has essentially shut down.

“So,” I said to my interlocutor, as I eyed the chocolate cake, “maybe this is not your vision of what a mother-daughter relationship should look like; maybe you don’t like the economy of errors in this family.”

“That’s close,” my interlocutor said, “but actually this is what I think I  is really going on in that house.”

Then they took my crayon and drew this (Figure 5):

Figure 5

“What’s this?” I asked.

“I think that Amy Chua is just shoving her theory of parenting down her daughter’s throat, and her daughter isn’t being given any space to say how she feels about it. Amy Chua is resolving all of her error by asking her daughter to change, but isn’t taking any of her daughter’s error in return.  It’s a command economy.”

They took the crayon out of my hand. “This is what I’m going for in my family,”  they said, and drew this (Figure 6).

Figure 6

“An economy of  errors” I said.

“Yes,” they said.

“And economommy of errors?” I ventured.

“Boo,” they said, “boo.”

And with that they left me to manage my error all on my own. I have survived; they were, for me that night, at least, my Tiger friend.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s