DSM Diagnoses are Families, Not Definitions

Labels like "autism" or "ADHD" can help people if used with understanding

The general argument of this post appeared several years ago on Blogspot. It has been substantially rewritten. Here it is, at last.

Did you know that there is an animal called an okapi?

An okapi is a mammal.

Based on that information, can you guess the answers to these questions?

1.     Does an okapi have fur, feathers, or scales?

2.     Does an okapi have a heart? A brain? A spinal cord?

3.     Does an okapi lay eggs or give birth to live young? Does it forage for food for its babies or secrete milk to feed them?

Photo of an okapi from the Website AZ Animals. © Raul654 / Creative Commons — License (Original)

Did you say:

1) fur;

2) yes, yes, and yes; and

3) gives birth and feeds young milk?

If so, you were right!

Have you ever heard of an okapi before? Did you have to guess the answers to these questions?

If so, congratulations! You’ve just demonstrated the power of categorical inference!

The power and perils of categorical inference

Categorical inference about members of a category gives us at least tentative knowledge about things we’ve never seen and might not even have heard of before.

However, we often make inaccurate inferences when we assume that the traits of a human group apply to an individual from that group.

We are especially likely to make errors when thinking about Diagnostic and Statistical Manual (DSM) categories, like autism spectrum disorder (ASD) or ADHD.

[Note: The name of the diagnosis is “autism spectrum disorder” or “ASD.” However, autistic people themselves use “autism” as a less pejorative shorthand. Thus, unless otherwise stated, I use these terms mostly interchangeably. However, I favor ASD when talking about the DSM diagnosis and autism when talking about individual people].

First, we assume that if a trait is typical of the group, then it must be true of a specific individual in the group.

For example, if men, in general, are over 5’6” tall, then a specific man, Bobby, must be over 5’6” tall. (Bobby is 5’5”).

Second, we assume that if an individual has a trait that is atypical of the group, that person must not belong to the group.

For example, if Bobby has many friends, and autistic people in general don’t have many friends, then Bobby must not be autistic. (In fact, Bobby is autistic and does have lots of friends — who are also autistic and share his communication style).

When it comes to protected categories (gender, sex, race/ethnicity, religion, etc.), it becomes more obvious that these are stereotypes. However, this way of thinking is stereotyping regardless of the group involved.

We make these inaccurate inferences because we think DSM categories work like “okapi” and “mammal.”

In fact, DSM diagnoses follow a different set of rules: they are “family resemblance” categories.

How family resemblance works

In the case of a category like “mammal,” the category is defined by having certain features, in contrast to other categories, which do not include these features.

For example, by definition, a mammal is warm-blooded, feeds its young by secreting milk, and has skin more or less covered with hair, not scales or feathers. Meanwhile, these traits are not true of animals in other classes, such as reptiles. [1]

So, if you pick any mammal — including an unfamiliar one, like an okapi—you can be sure that it is warm blooded, nourishes its young by secreting milk, and has hair or fur. Pick any animal that is cold blooded, does not secrete milk, and has scales or feathers instead of hair, and you can be certain it’s not a mammal.

A family resemblance category doesn’t work that way. There are no traits that are true of every individual group member and untrue of everyone outside the group. However, certain traits are very common in that group or rare in other groups, which distinguishes group members from non-group members. The difference is big enough to make the category useful and meaningful.

Philosopher and literary critic Ludwig Wittgenstein, shown above, pioneered the term “family resemblance,” using the example of games. What single characteristic is true ONLY for every “game,” including sports (basketball), board games (chess), card games, video games, trash talking, playground games (hopscotch), and word games (riddles)? He argues there is no such characteristic, just as there is no physical feature that is true of every member of a human family, and no other families. Yet, we can easily see traits that most members of a family have, and most other people do not. “Games,” and families, are still meaningful, useful categories.

For example, one might say that “lions have manes.” In fact, only adult male lions have manes. However, having a mane does distinguish a lion from other kinds of big cat.

Similarly, DSM diagnostic categories are defined by differences from the rest of the population — those who are not impaired by specific traits.

For example, a depressed person takes less pleasure in things, feels less hopeful, and has less energy than someone who is not depressed.

DSM diagnostic categories are often based on clusters of traits that occur together more often than you’d expect by chance. People invent such groups easily and often. Consider the following examples, arranged in order of increasing scientific support and mechanistic basis:

·        Aquarius (astrology);

·        INFP (Meyers-Briggs personality system and its variants);

·        Introvert (many personality psychology theories);

·        Inattentive ADHD (DSM-IV and DSM-5).

Autism as a family resemblance category

In the United States during the 1940s, Leo Kanner noticed children with several co-occurring traits. Each trait was rare in the general population, and the combination was extremely rare.

In Germany under the Nazi regime, Hans Asperger identified children with a similar set of co-occurring traits. Each of these traits was rare in the general population and the combination was extremely rare.

What these men named “autism” and “Aspergers syndrome” were clusters of co-occurring traits. And today, (according to DSM-5), they are part of the same “family”: the children both Kanner and Asperger observed would all be diagnosed with “autism spectrum disorder.”

ASD is a fantastic example of family resemblance. Two people correctly diagnosed with ASD can be even more different from each other than they are from the average person. For example, some people diagnosed with ASD are excellent public speakers, while others can’t speak communicatively at all.

Moreover, there are many variations within this “family.” 2 children can be diagnosed with ASD while having no diagnostic traits in common! In the previous DSM edition, DSM-IV, some 2361 combinations of traits led to an ASD diagnosis.

Autism is a family resemblance category, and should not be treated like a definition.

We can’t make generalizations about autistic individuals the way we can about different species of mammals.

That distinction sounds abstract, but it has very real effects on how we think about, talk about and treat people.

3 ways it hurts neurodivergent people when we make false assumptions from their diagnosis

When we fail to understand that DSM categories are based on family resemblance, we stereotype neurodivergent people in several damaging ways.

1. We meet the needs of the person we expect, not the real person

First, we assume that traits that are common among people with a diagnosis are true of all people with the diagnosis.

For example, a teacher hears that a new student with ADHD, Alex, will join the class. The teacher may assume that Alex is:

·        male,

·        hyperactive (has hyperactive or combined type ADHD),

·        will disrupt the class,

·        will have difficulty with reading or math,

·        will have a short attention span whether interested or not,

·        will excel at multitasking and have difficulty focusing on a single thing.

In fact, Alex turns out to be:

·        a girl,

·        not hyperactive (has inattentive type ADHD),

·        daydreams instead of disrupting the class,

·        excels academically when interested,

·        can focus for hours when interested,

·        has difficulty multitasking.

Yes, the teacher is using outdated stereotypes about ADHD. However, the underlying problem is that the teacher thinks that traits that are true of many people with ADHD must be true of Alex, specifically. That error won’t be fixed just by providing more information about ADHD.

Plus, the teacher isn’t entirely wrong. Hyperactivity really is common among people with 2 out of 3 types of ADHD, and less common among people without ADHD. However, Alex isn’t hyperactive.

People with ADHD are disproportionately likely to have reading and math disabilities and underachieve in school. That doesn’t mean Alex does.

If teachers assume Alex won’t excel at math or reading, they might not offer sufficiently challenging work.

If teachers assume Alex will disrupt the class and she daydreams instead, they may ignore her falling ever further behind.

It’s especially damaging when teachers are the ones with false assumptions, because students tend to live up or down to their teachers’ expectations.

2. We assume people with different diagnoses do things for different reasons

When people with different diagnostic labels do the same thing, we assume that they must be doing it for different reasons.

Suppose a child doesn’t pay attention in class. If the child is labeled “gifted,” we assume they are bored and unchallenged. If the child is labeled “ADHD,” we assume they have difficulty paying attention. In fact:

·        A gifted child can have difficulty paying attention.

·        A child with ADHD can be bored and uninterested in the subject. In fact, a child with ADHD might be especially prone to boredom. Everyone has difficulty focusing when not interested, but it’s especially hard for children with ADHD.

·        The class can move too slowly, and schoolwork can be too easy, for a child with ADHD.

·        The class might actually be boring (to anyone, of any neurotype).

·        There could be some other explanation unrelated to a child’s diagnosis.

If we make assumptions based on a child’s diagnosis and don’t consider other possibilities, we may misunderstand the child and fail to address the real problem.

3. We deny their diagnosis and their support systems

When someone does not have a trait we associate with a diagnosis, we assume they don’t have the diagnosis, or at least question it. That error is especially insidious when neurodivergent people do it to themselves.

On Tumblr, Twitter, and Quora, I often see people ask, “I don’t do/have/feel x. Am I still autistic?” For example, “I’m not good at math. Can I still be autistic?” (The answer, by the way, is almost always yes, it’s possible).

Often, the people asking seem to be confused, hurting, and doubting their own judgment. They seem to question their worthiness to belong or to connect with others who share their experiences.

Also, when we make false assumptions about people with disabilities, we force them to spend time and energy correcting our misperceptions. They don’t need and can’t afford that extra burden.

We should use diagnostic labels carefully, not abandon them

This frequently-used phrase is shown here on a t-shirt sold on Redbuubble by clareecrafts.

Some people believe that labels are always harmful, saying things like, “labels are for soup cans, not people.” They do not seek a diagnostic evaluation for their children. If the child does receive a diagnosis, they do not inform the child.

These parents see children with DSM diagnoses being siloed away from mainstream classes, burdened with low expectations, and bullied by peers. These parents want to protect their own children from such a fate.

The problem is, these parents seem to assume that if children go unlabeled, they’ll be just like other kids, and others will treat them just as well.

In reality, neurodivergent kids still think and act differently. When they can’t do something expected of them, people assume that they can do it, but won’t.

A different set of labels dogs them, bringing along a different set of negative prophecies: “uncaring,” “lazy,” “selfish,” “difficult,” “overly sensitive,” “rebellious,” etc.

Neurodivergent kids know they’re different, but have no idea what’s “wrong with them” or what to do about it. It’s a bit like growing up missing the operating manual and having to reconstruct it yourself, without help.

Jean Gray writes, “Being an undiagnosed neurodivergent individual is inherently traumatic.” I agree.

People diagnosed as adults usually experience a sense of relief and understanding. “So that’s why I’m always tired” or “that’s why I don’t keep jobs for very long.”

The diagnosis gives them a new story for making sense of their lives, and gives them perspective on key events and relationships in their lives. It allows them to let go of self-blame and shame. Along with relief comes grief and sometimes anger at lost years and opportunities.

A diagnostic label also comes with a community: other people who bear the same label, share experiences, and deal with similar challenges. Newly diagnosed adults often fling themselves into autistic and ADHD communities where they feel understood, sometimes for the first time in their lives.

Not to mention the services and legal protection that people with a disability label can access.

The cost of “throwing out the labels” isn’t worth it.

Instead, we need to stop misusing them.

When we understand that these labels are based on family resemblance, we can discuss people with disabilities with care and curiosity, not stereotyping.

Footnotes

[1] “WELL, ACTUALLY…” It’s slightly more complicated. Birds (class aves) are also warm-blooded, so an animal could have warm blood and not be a mammal. However, birds don’t have the other traits. Thus, it’s really the combination of multiple traits from this list that makes an animal a mammal. That’s not important. A more complicated definition is still a definition.