Day 3: Psychological Jade and The Arrogance of Ignorance

Introduction
Before reading this, I suggest reading my post from yesterday since most of what I talk about here relates to it (and I end up retracting most of what I said).

'The arrogance of ignorance' is one of my favorite phrases. I'm not sure of its origins but I heard it first from Dr. Steven Novella. I think the phrase is the best way to capture a cluster of common cognitive errors. In no particular order: a) Assuming that because you are knowledgable in one domain that you know a lot about another (or are able to correctly evaluate another). b) Moving from small data sets/anecdotal experiences to broad conclusions. 'The arrogance of ignorance' is a close cousin to the Dunning-Kruger effect: You have so little knowledge of a particular domain that you are unable to assess how little you actually know and grossly overestimate how much you do know, in turn leading you to wildly wrong conclusions.

Anyhow, yesterday I was guilty of all of the above crimes. You'd think a guest pass to a hospital and a few hours of observation would give me enough authority and knowledge to correctly evaluate an entire subfield of medicine. Strangely, it didn't.

Today I went back to inpatient psych, and boy am I glad I did (from a pedagogical point of view). Let me try to both convey my experience and undo some of the misconceptions I had.

Sample bias
Most of the patients I met yesterday had been in the hospital for over a week. I was meeting them after they'd undergone treatment and had been stabilized. Of course they seemed normal to me! As I learned (and hopefully you will too once you read this post), what I did was the equivalent of walking into a surgery unit and looking only at the patients about to be discharged, then asking"why did they need surgery? They look fine to me!"  

So, what are patients like on admission and early in treatment?

Obviously there are a variety of disorders but all of them are severe. Here's the thing, unless you work in a hospital or have someone in your family with a severe mental illness you've probably never actually seen severe mental illness. To most of us, this is an invisible population because most of their lives are lived in care homes or in institutions and, unfortunately, in the streets.

The patients range from very well-spoken with linear thought to having only elementary vocabulary with disjointed unintelligible thought, and any combination of the above. Regardless of where they fall on the spectrum, most of them suffer from severe delusions. Some examples: (1) Being part of an intergalactic group of assassins being pursued by the (intergalactic) mafia, (2) believing that a family member is dead who isn't and all the evidence they have should lead them to conclude the opposite, (3) being pursued by terrorists and (actually) destroying windows and cars to avoid/prevent the terrorist plot, (4) having voices in their head telling them to kill others or kill themselves. There were more but this should suffice.

Interestingly, the ones that had grand conspiracy delusions, e.g., (1) and (2), were extremely pleasant to talk to. If you were to have a conversation with them and the content of the delusion never came up, you wouldn't suspect a thing. It was as though they simultaneously inhabit two realities. When you ask them, they know where they are and why they're there. They'll say "I just want to get better" but at the same time they'll discuss their delusions as though they're just as real as the chair they're sitting on.

Unlike what I hypothesized yesterday, these people don't "just need a little more social and material support." That's the equivalent of saying someone with cancer can be healed with a back rub.

I think my reaction yesterday is probably analogous to what happens with many deniers of modern medicine. They've been in the hospital to visit a friend or they read some article online--maybe even spoken to a disgruntled doctor.  But they've only seen 1 billionth of the data set, and only from the patient side of the bed. Things look very different from the doctor's side of the bed and as you get a larger data set...

Tip of the Iceberg
Another factor that led me to my (wayward) conclusions yesterday was I didn't ask enough questions about case histories. Once you read the case histories, your perspective will change very quickly. Every patient in there has a lifetime history of psychosis that is well documented. Almost all have been suffering the same symptoms since adolescence. Some have their condition for unknown or unknown biological reasons (usually genetic, as it runs in their families), others (there were 2 there) had suffered major brain injuries at some point earlier in their lives and haven't been the same since, others have their condition as a consequence of a life-time of substance abuse. For many it was a combination.

Someone who thinks that a little positive thinking or mere talk-therapy is going to solve these people's problems is extremely naive--like I was yesterday. Someone who thinks along these lines is mistaking people who have one-off breakdowns or depression with this other population. Like I said, unless you work in a hospital or have a family member (or work in law enforcement, probably) it's unlikely you've ever met anyone from this unfortunate population. Until you do, you can't fathom just how serious it is.

Philosophy of Science Lesson: Depression and Jade (Bear with me, You'll See How this Relates in a Moment)
What is jade? Up until the 19th century it was believed to be a kind of mineral. However, a French mineralogist (Alexis Damour) discovered that it was in fact two distinct minerals: Jadeite and nephrite, each with distinct chemical and structural properties. Nephrite is a microcrystalline interlocking fibrous matrix of the calcium, magnesium-iron rich amphibole mineral series tremolite (calcium-magnesium)-ferroactinolite (calcium-magnesium-iron). Jadeite is a sodium- and aluminium-rich pyroxene. The gem form of the mineral is a microcrystalline interlocking crystal matrix.

This isn't Rocks for Jocks 101, so why should we care? I'm getting there. Notice that both structurally and chemically, nephrite and jadeite are different. What does this mean? Well, we know that different chemical compositions will react differently and different microstructures will also also behave differently. Jadeite and nephrite have different fundament properties. From the point of view of science, if we think that science divides and studies the world in terms of its fundamental rather than superficial properties there's no such thing as jade. There's no one structure or chemical structure that is jade.

Here's another way to illustrate what I'm getting at. Why isn't there a science of green things? Why aren't there green-ologists? The reason is that green is a superficial property. Knowing that something is green gives us no predictive power in terms of how other green things will behave. It also provides no explanatory power for why it behaves the way it does. 

For example, green algae has very different fundamentals chemical properties from a green glass. Suppose I put a HCL on the green algae. Based on the chemical reaction, would I be able to predict what will happen if I put HCL on green glass? Does the algae's greenness explain why it reacts the way it does to HCL? Of course not. In science, I want to lump things into categories that are going to allow me to make generalizations and predictions about other things in that category.  

Learning about green algae doesn't help me learn anything important about green glass except what I already knew--they're both green. We don't lump the two into a scientific category because science is only concerned with "lumping" things in terms of shared fundamental properties rather than superficial properties. We ought to "split" superficial categories that contain objects that have different fundamental properties.

Ok, so what does all this have to do with psychiatry and psychiatric diagnoses?  Consider that a common diagnosis we hear about is depression. Most cases we (by we, I mean non-medical professionals) have encountered are probably infrequent non-pathological affairs, perhaps set off by a traumatic event. We know that, with support, most people eventually work through the depression and end up fine. The problem is, 'depression' is psychological jade. 

Different types of depression can manifest the same superficial symptoms but the underlying causal structures are different. (No, alt-meders, this isn't the same "root cause" you're thinking of but it's the one you ignore).  So, the mistake is to think, "ah, depression...we just need to treat it with x, that's what we did with the last case". But this is to treat depression like jade--i.e., as a homogenous category based only on superficial resemblance.

For example, I learned (the very surprising fact) that for many types of deep depression the most effective treatment is ECT (electroconvulsive therapy)--yes, you read that right! I had to ask the doctor twice because I couldn't believe my ears. Apparently, it's well studied. Of course, the current procedure is quite different from how it was in the early days but still...who'da thunk? 

"Jadists" about depression might think all cases of depression can be treated with ECT. This would be a mistake. There are different kinds of depression with different etiologies (underlying causal structures). It turns out that depression in manic depressives doesn't respond to ECT. Depression has its own jadite and nephrite (and more). The "root cause" of depression in manic depression is fundamentally different than it is in other kinds of depression.

Beside this overview of a famous philosophical argument, why am I talking about this? Because if you're a human being you're probably going to commit the same cognitive error that I made when it comes to psychological diagnoses. You hear that a patient (or someone you know) is depressed or has some other general psychological problem and you think about how you or someone you know dealt with it. You think, well, all they have to do is x (whatever worked for you or your friend). It's so simple! 

But you're treating the diagnosis like psychological jade. The diagnosis might have the same symptoms but it doesn't mean it has the same underlying fundamental structure and thus, there is no reason to suppose it will respond to the same intervention. It's a different kind. 

Worse yet, someone could dogmatically claim the all disease and/or psychological diagnoses share the the same "root cause". Such dogmatism precludes any chance of recovery since the same ineffective treatment will only be applied more and more vigorously. What's more, this way of thinking is the opposite of scientific thinking. Saying everything has the same "root cause" is just like being a green-ologist. You're confusing superficial similarity for fundamental similarity. To use the lingo of metaphysics, you're lumping when you should be splitting. 

We see green-ology all over the place in alt-med. For chiropractors, the "root cause" of all disease is some sort of spinal misalignment, for Ayurvedic medicine the "root cause" is chakra alignment (or some shit), for reflexologists the "root cause" is something to do with your feet (WTF? How are these people even a thing?), etc... (While I'm pointing the finger I should make clear that I have my own "root cause" default. I have a tendency to lump various problems as being caused a general lack of meaningful social relationships, belonging to a community, and sense of purpose.) And then there's alt-med's favorite: stress. The "root cause" of all disease--physical and mental is stress. More green-ology. 

To be charitable we can say that stress can trigger or make people more susceptible to disease but this is to confuse notions of causation. Let me illustrate. Suppose someone is in the hospital with a broken leg because they got hit by a car. What "caused" the broken leg? Being hit by a car, right? Now, just because the car was the trigger for the broken leg no one in their right mind would believe that removing the car will heal the leg. 

I can just imagine the doctors at an all-alt-med hospital: "We've cured your leg by getting rid of the 'root cause'--the car has been destroyed! You can walk now!"

So, while it's true that stress can trigger certain reactions, it doesn't follow that the solution to the problem is merely to remove the trigger. Yes, doing so may decrease the likelihood of the same event from occurring again, just like not getting hit by a car will prevent you from breaking your leg again (that way); however, this insight is often of trivial value. No one with more than two brain cells to rub together thinks chronic stress is good for them. What's the next great insight? A poor diet isn't good for your health? Revolutionary! Please collect your Nobel Prize.

The Lesson
The causes of many diseases, physical and mental, have to do with their fundamental underlying structural properties. This is why people respond differently to different treatments. Superficial similarities can cause us to lump when we really should be splitting. Overzealous lumping leads to failed treatment and frustrated patients. Overzealous lumpers are green-ologists. Don't be a green-ologist. 

Anyhow, this is just one more cautionary tale for me to heed. Hopefully, it gives you pause too the next time you diagnose someone (including yourself) and assume that superficial similarity implies fundamental similarity...

Also, hopefully this little digression shows the value of philosophy to science. You can't do one without the other.

In a Nutshell
The conclusions I drew in my last post were wrong. But I'm leaving that post up as a cautionary tale to both myself and to anyone reading this. My hope is that it reminds us how easily we can get things wrong when we only have a little bit of information, particularly about areas where we are not experts. People think it's "being a sheeple" to defer to experts. It's not. It's smart and good epistemic practice. Only arrogance fueled by ignorance would lead a person to think that they know more than an expert in that expert's domain.