[content warning: mild ideohazards about rumination that might make people who have anxiety disorders have anxiety disorders more effectively, in the bad sense of “effectively”]

[epistemic status: more crackpottish than usual for this blog. Wild speculation.]

I.

If the brain had been designed by an amateur, it would enter a runaway feedback loop the first time it felt an emotion.

Think about it. You see a butterfly. This makes you happy. Being happy is an unexpected pleasant surprise. Now you’re happy that you’re happy. This makes you extra happy. Being extra happy is awesome! This makes you extra extra happy. And so on to as much bliss as your neurons are capable of representing.

Or you stub your toe. This makes you sad. Being sad sucks and makes you less productive and less fun to be around. This means being sad is an unexpected negative event, which sounds like a good reason to be sadder. Being even sadder is going to ruin your night even more, so you get sadder still, and so on until you become suicidally depressed over a single lousy toe.

In the real world, either those feedback loops usually don’t happen, or they converge and stop at some finite point. I would not be surprised to learn that a lot of evolutionary innovation and biochemical complexity goes into creating a strong barrier against conditioning on your own internal experience.

Sometimes it fails. A guy named Wegner conducted a famous psych experiment where he asked a bunch of participants to sit in a room alone and try not to think about a bear with sunglasses. Of course, that was pretty much all they could think of. They seem to have gotten into a feedback loop where “desire not to think of a bear with sunglasses” -> “thought of a bear with sunglasses” -> “frustration” -> “stronger desire not to think of…” -> “more thoughts of…” and so on.

One of my professor’s work – and my college thesis – expanded the field of bear-with-sunglasses-ology to note that people with obsessive compulsive disorder are found to be much worse at this task than the general population. This is not surprising. OCD seems to be pretty much exactly this scenario, except instead of trying not to think of a bear with sunglasses, you’re trying not to think about how you’re dirty-contaminated, or how maybe you left the stove on, or how what if your car just hit somebody then where would you be? OCDers get stuck in feedback loops where their worry about their obsessive thought is itself a form of the obsessive thought and so both justifies and intensifies their worry.

I also remember a study where a guy with a sleep lab offered participants a cash reward if they could go to sleep more quickly than usual; the end result was that all of them took much longer to go to sleep than usual. Strong desire to go to sleep -> actually a strong anti-sedating emotion -> worry that you won’t get the prize -> stronger desire to go to sleep -> even less sedated -> so on.

There seems to be an element of this in most anxiety disorders. Someone goes outside, something bad happens. Next time they go outside, they feel anxious. The usual strong barrier against conditioning upon internal experience is AWOL for some reason. The patient finds the experience of becoming anxiety very negative; therefore their belief that “going outside leads to bad things” is justified. Eventually they are so anxious about possibly becoming anxious that they just stay in their house all the time.

(hey, this is kind of Lob’s theorem! If you know that, if you’re anxious about being anxious you would be anxious, then you’re anxious. Maybe.)

But the most clear-cut example is panic disorder. Someone gets anxious for some reason. They get the standard somatic correlates of anxiety – racing heartbeat, sweating, etc. This is a scary situation to be in, not least because it mimics all sorts of terrible medical conditions like heart attacks. This makes the person more anxious, which increases the somatic correlates, and so on.

II.

A lot of CBT seems to be about manually breaking these feedback loops. But I’m just as interested in what we accidentally do to manually increase them. I worry that making mental self-reference slightly easier and more mentally accessible can lead to a big increase in feedback loop size.

I am misophonic – it means I can’t tolerate certain noises. Ten years ago, I would not have used the word “misophonic”, and I would not even have said I have low noise tolerance. I would have said “Hey, that TV is bothering me, can you turn it off?”

I didn’t start thinking about it on a meta-level until one of my roommates told me “Wow, Scott, you seem to be really super sensitive to noise.” And then gave me a little bit of grief over it, which made it stick in my brain. Ever after that I modeled myself as a person who was super sensitive to noise.

And that made my noise sensitivity much, much worse. I hypothesize that maybe, instead of just noise -> distraction, this created a longer feedback loop. Something like noise -> anxiety that I, as a person sensitive to noise, am going to be distracted -> this anxiety is itself distracting -> noticing that I am distracted and being anxious that this distraction will continue as long as the noise continues -> further distraction -> and so on.

(it’s worth noting here that I have obsessive compulsive disorder and that noise sensitivity is a classic feature of the condition)

I had precisely the same experience on this post when I said I was triggered by certain kinds of feminist and social justice rhetoric. I didn’t really think of it that way until I wrote that sentence. I mean, it was true that I heard the rhetoric and then I felt upset and scared, it was just that I had never thought of it as a part of my identity before, or connected it to the word “trigger”. Well, as soon as I did that, the problem got about three times worse, and continues to get worse. I think this nice little crystallized concept of “trigger” might allow my brain to feed back its anxiety more effectively, like “Yup, this thing triggers you, better start feeling anxious about feeling anxious about feeling anxious about…” and that made it go from “a thing that bothers me but which I can cope with” to “giant psychological disaster”.

(now I wonder about typical mind stuff. Do people without OCD have this same experience of anxiety about anxiety about anxiety…? For example, when I was young I was afraid of the dark, not because I believed in ghosts, but because I expected to hear a sound or see something blown around by the wind which I would mistake for a ghost and then have to deal with waves of terror rushing through my body until I figured out what it was, and this was sufficiently bad that I slept with the lights on as a child. Is this the sort of thing other people could imagine feeling, or is it really unusual?)

But this sort of thing is comparatively small fish. I’m more worried about the effect of our entire rich emotional vocabulary. Like, we just sort of invented the concept of being “stressed out” sometime in the last century. Did that make people more stressed out about being stressed out? Did the movement for people to become more introspective and talk about their feelings more (that was a thing, right? That’s why old people are so stoic and young people are so touchy-feely?) make people more likely to fall into feedback loops?

What happens when you give people a psychological diagnosis like “depression”? I’ve always heard that it makes people feel better, because now they know there’s an explanation for what they’re experiencing and it’s not their own fault. But I don’t know if I’ve seen any studies proving this, and even if there were I’d expect them to suffer from the general bias to confirm things that everyone knows are true. What if it just makes people be depressed about how depressed they are, and then go “There’s that depression again, guess this means I’m not getting any better” and become depressed about that?

III.

This whole essay is a little crackpottish, but now I’m moving from things that merely can’t yet be supported by evidence to things that actively contradict it. But I think about this a lot, and it’s my blog, so shut up and listen anyway.

I notice that the class of mental disorders that seem to involve feedback loops – depression, OCD, anxiety – are also the class of mental disorders effectively treated with serotonergic drugs.

One of the most powerful serotonergic drugs in existence is LSD. And as I have learned from – let’s say long boring journal articles – the main effect of LSD is to make you literally loopy. Your thoughts loop in and become about themselves, your sense of time becomes cyclic, everything you see becomes a fractal or a spiral or both. And you end up in an extreme emotional feedback loop to infinity – either a “good trip” or a “bad trip”.

This just-so story is not quite as convincing as I would like because SSRIs and LSDs both increase serotonin but have opposite effects on loopiness. But “serotonin” is a wide and complicated category, and like I said above, I bet the brain has a lot of different mechanisms to finely adjust how self-referential your thoughts are allowed to get.

I will get crackpottier still: maybe the parameter being adjusted is some kind of “allowed size of loop”, so that usually you can finish an entire train of thought and then maybe reflect back on items in the train. Small amounts of LSD decrease the loop size, so that individual thoughts can refer to themselves. And large amounts of LSD (the journal articles I read were very comprehensive) decrease the loop size to zero, and the perfect pure consciousness people claim to experience is just a loop looping in on itself forever, empty of content.