The Mislabeled Child: On Autism Spectrum Disorders (ASD)

This is second installment of my discussion of The Mislabeled Child. Here is a link to the first installment. This is a long one, so I've divided it into parts.

I Background

As I blogged, the authors write through a "neurolearning" lens. Of labeling, they write:

Labels matter becasue they can cause us to take an overly sterile, detached, and clinical view of a child's learning challenges and forget that these challenges occur in the context of a child's unique and complex life. They can make us overlook the important role that a child's unique temperament plays in dealing with learning challenges, and our need to help children develop character traits like optimism and resilience. [Emphasis added.]

The problem is: science is sterile and detached, by definition. And arguably, those very qualities are what provide a diagnosis with integrity and rid it of bias. Education and learning are, at best, social sciences and at worst, entirely unscientific. Autism, by contrast, is a biological condition - a life science, not a social science.

II Mislabeled

Applying the neurolearning lens, the authors conclude that autism is a commonly misapplied label. These mislabeled children, they argue, often have other conditions that are superficially similar to autism. They write:

...[c]hildren with these other conditions don't show the severe deficits in emotional empathy, "mirroring," and social affiliation (or sense of bonding with others) characteristic of autistic disorders.

They also report that children with autism present with different brain scans when compared to children with "other conditions." From a learning perspective, the group the authors label as autistic are generally impaired with respect to higher order thinking.

Noticeably absent from the differential diagnosis: engagement, the lack of which is commonly regarded as a defining quality of autism. They concede that these mislabeled children often have difficulty with social interactions and are superficially similar to autistic childen.

In essence, the authors are drawing a line where it doesn't otherwise exist from the perspective of empirical observation of behaviors (which they pejoratively describe as clinical and sterile). They don't expressly argue that practitioners are misapplying the DSM-IV diagnostic criteria. Rather it appears that they are arguing that, to get at the essence of autism, we need to go behind the behaviors. Pop open the metaphorical hood and poke around.

III The Good

What works best about the neurolearning paradigm is the prescription for treatment and particularly, how to teach or apply learning strategies to autistic children. This is a pet issue of mine, right? I started blogging in part because Brad's doctor was pushing ABA and I've faithfully blogged against ABA-pushing ever since. The Mislabeled Child articulates a great, easy-to-understand prescription for treatment for autistic children and nonautistic children alike. What follows is an outline of the relevant section on autism:

General Principles for Helping Children with Autistic Disorders

• Reward and Motivation
• Using Their Best Learning Strengths and Styles
• Simplifying Information Input and Pattern Processing

Helping Children with Autistic Social Impairments

• Helping Autistic Children to Develop Empathy
• Helping Autistic Children Improve Their Social Interactions

Helping Children with Autistic Communication Impairments

• Improving Verbal Comprehension
• Improving Nonverbal Comprehension
• Improving Verbal Expression
• Improving Nonverbal Expression

Helping Autistic Children with Attention and Self-Regulation

• Helping Autistic Children Improve Attention
• Helping Autistic Children Improve Emotional Self-Regulation

Their roadmap for treatment is helpful, and I would encourage you to buy the book if you are interested in learning more.

The authors don't directly prescribe ABA as a treatment, but rather intimate that some of the underlying principles (reward and motivation) have merit. They write:

...[a]utistic children typically have a diminished desire to imitate others or to please and earn praise from others. In this way, they differ from most children, who quite naturally want to imitate and please parents and other adult authority figures - like teachers. Usually we take these desires to imitate and please others for granted in designing our educational and therapeutic systems. However, when a child has little intrinsic motivation to imitate or please, we must find a way to motivate her using things she finds extrinsically rewarding. [Emphasis added.]

The authors don't go on to draw the negative implication, so I will: if a child has the desire to imitate or please, it doesn't make sense to use an intensive program of therapy like ABA which is based on a system of extrinsic rewards. Common sense, right?

Not so much. As I blogged a few weeks ago, Deborah Fein et al published a study in which a group of children who present with "verbal and motor imitation" skills were said to recover from autism, and the study implies (but does not prove) that intensive ABA therapy caused this positive outcome. So there's a group of respected researchers who have taken common sense and turned it on its head: let's take the children who are imitating (without extrinsic rewards) and subject them to an intensive system of therapy (40 hours per week, recommended!) based on extrinsic rewards.

IV The Shortcoming

While I think the neurolearning paradigm is compelling, it does have a shortcoming: it's ahead of its time, for two reasons:

1. The neurological indications (ie abnormal brain scan) are not dispositive as biological markers for autism at this juncture, from a scientific perspective. There are clusters of patterns of statistical significance, but not enough to define autism in whole, because autism itself is such a heterogenous condition. For example, from an etiological perspective, there may be five different autisms, three of which have observable neurological indications (ie abnormal brain scan) and two of which do not. In truth I don't know how many etiologically distinct "autisms" there are; no one does. Today. Maybe tomorrow, but not today. And even if we fast forward and hypothesize that biological markers will be discovered, who's to say that they will take on the form of neurological indications? Perhaps, one day the "litmus test" for autism will be a genetic test, not a brain scan.
   


2. The correlary to autism as a mislabel is that these children often have an "other condition" that mimics autism. In other words, it's "you don't have this (autism), you have THAT." However, at least one of the THATs - sensory processing disorder - isn't recognized in the DSM-IV. This means there is no uniform way of diagnosing it. It also means that the label won't be respected or understood by the public school systems. In this blogger's opinion, a label is useless and perhaps harmful if it isn't uniformly applied or understood. Which, ironically, is the argument commonly used against the liberal application of the autism label. But the same argument can be made, more forcefully, with respect to SPD. (To be clear, I'm not arguing that SPD is a bad diagnosis or doesn't exist; I'm merely arguing that, today, it's poorly understood and not respected.)

More on sensory processing disorder and where I think Brad fits into it all in my next installment.