A few whimsical thoughts on consciousness:

-Evidence suggests that consciousness is integral to the resolution of conflicting motor commands (think how intensely aware you are when you have to hold on to a burning plate, or when you’re pulling off a band-aid, or any other time you have to fight a reflex)

-Consciousness seems to be tied in with executive function (they’re both located near and around the frontal cortex), which is also important to the regulation of conflicting motor commands.

-Various parts of the brain have been mapped specifically to restraint. The ACC is often associated with ‘conflict monitoring’, and the DLPFC is associated with sorting out conflicting information. When we’re asleep, the DLPFC is shut down, which is why we cannot notice the ridiculous dreamworld generated for us while we sleep. Another brain area, the DFMC, shows activation when an action is inhibited.

-Frontal lobe damage results in a loss of impulse control, and children (who have less developed frontal lobes) tend to be notoriously impulsive.

-The frontal cortex is likely the last brain area to have evolved (as it is the last brain area to grow in babies). It was likely added on to an already adequately functioning brain, one that was simply following basic emotional scripts. Conscious thought is a slow process, and in times of stress we rely on the quick judgments of our emotional system instead.

-The infamous Benjamin Libet experiments show we’re not conscious of an action until we’ve already decided to do it.

All the evidence points to consciousness serving an inhibitory function, essentially neurological veto power. This is interesting enough by itself, but what really intrigues me is how this function might relate to how the brain learns (specifically, how it approximates Bayesian updating and/or hypothesis and test).

One of the most useful manufacturing concepts ever invented was the idea of interchangeable parts. It was (depending on who you ask) first used by Eli Whitney for the manufacture of rifles. Before interchangeable parts, each rifle was essentially a custom job, built by hand. The main pieces would be largely the same, but there was no guarantee that a trigger on one rifle would fit properly on another, or that the barrel of one would rest properly in the stock of another, etc. As a result, guns were slow to produce and to repair.

The money idea behind interchangeable parts is standardization. And the benefit of standardization is that it removes thought and judgment from the process. Instead of figuring out how to repair a rifle, you just remove the broken part and put in an identical one. Instead of having one assembled by a knowledgeable craftsman, you can give the basic instructions and the pile of parts to unskill laborers. Instead of having each piece carved by hand, you can have hundreds pumped out by dumb machines. Standardization takes intelligence out of the process of creation and moves it back a level, into deciding how the process of creation should work. Once the design of a rifle is completed, and the proper castings are forged, everything that remains is a dumb process, provided things run smoothly. You know longer need to understand how a gun works to be able to repair or assemble one.

So perhaps it’s more accurate to say that rather than intelligence being removed, standardized parts come with intelligence preinstalled.

Semantic satiation is what happens when you repeat a word so many times that it seems to lose all meaning. The process behind this is something called ‘reactive inhibition’ – essentially, your brain gets bored with the word and produces a weaker response each time the word is heard, eventually dropping off altogether. I am neither a psychologist or a neurologist, but on the surface this looks like an instance of nerves ceasing to fire when repeatedly exposed to the same stimulus. It’s the reason why you quickly get used to any sort of strange smell (and why you generally can’t smell yourself). The brain doesn’t want to waste it’s resources taking into account a constant feature of the environment, so once something seems to be there to stay, it gets ignored. Any stimulus repeated enough times will eventually fail to illicit a reaction at the neurological level, so it seems reasonable that these effects are visible at the psychological level.

Because of this, any constant feature of your environment will eventually be ignored. It’s this ability that lets us adapt to new situations, no matter how bad (or good) they might be. That horrible job doesn’t seem so horrible after a few weeks at it, and that solid gold house doesn’t seem so amazing after the 100th time you wake up inside it. We get used to things, for better or for worse – pleasure and pain fade over time. What doesn’t fade is desire – our reward circuitry is (I think) decoupled from this, providing a constant need to pursue more no matter how much we already have. This might be why it’s so easy to be greedy – we’re designed by evolution to pay attention to what we don’t have, not what we do.

If you’re the villain of an 80’s action movie, and you’re fighting the hero in the climactic showdown, there’s a good chance that you’ll try to throw sand (or dust, or powder) in the eyes of the hero to try to blind him. After all, he can’t fight if he can’t see, right? Obscuring the conflict should be great way to overcome your own lack of skill. It won’t work, unfortunately – the hero, though blinded, will still defeat you, making his victory and your defeat all the sweeter.

In real life, however, kicking up some dust is a great way to win, or at least to avoid losing. And just like in the movies, it’s a dirty trick that only villains use.

I am, of course, talking about arguments. Again. Specifically, I’m talking about kicking up dust by making things more complex than they need to be. Once any counter arguments become impossible to predict, impossible to check or impossible to know, you don’t have to worry about discarding your assumption – since you can’t prove it wrong, it must be right!

Example:
Woman: You should quit smoking, it’s bad for you.
Man: The studies that show that were all performed in the sixties with an extremely flawed methodology – they may have even been forged.
Woman: What? Why would you think that?
Man: You can’t be sure they WEREN’T forged, and the issue has become so politically charged that any current studies are suspect. People think it’s bad, and so they’ll try to interpret the evidence to show that it’s bad. The same thing happened with the concept of nuclear winter.
Woman: Yes, but…
Man: And there’s plenty of examples of conventional wisdom about medicine being way off base – look at all the hoopla about low fat diets. The human body is amazingly complex and we’re really not very good at teasing cause and effect relationships out of it.
Woman: Well…
Man: Besides, anything’s bad for you if you have too much of it, and its not clear how much you have to smoke before any alleged health problems start to show up. The science surrounding smoking is murky at best.
Woman: I want a divorce.

~fin

This sort of “it’s complicated, therefore it’s probably wrong, therefore I’m probably right” chicanery can be done for almost anything that can’t be directly verified. Much like arguing against basic principles, fog like this is difficult to penetrate. And also like arguing against basic principles, using this strategy makes you a dick.

Something I struggle with is conveying my exact state of knowledge about something. Much of the fault can be attributed to the brain’s design – the data my beliefs are based on is buried under miles of tangled subconscious tape. It only surfaces as an intuition, it’s original source having long been worn away. If I’m lucky, and it’s a topic I know extremely well, it will drag the necessary supporting beliefs along with it, and I’m capable of speaking somewhat intelligently on it. If I’m unlucky, it’ll instead bring strong emotions that further cloud the beliefs origins, and I end up looking the fool.

This is made worse by the fact that a state of ignorance is a state that I loath, regardless of how accurate it may be. “I want to be right” I think, and I proceed to take whatever steps my self-preservation instincts deem necessary. This includes (but is not limited too): overstating my case, deliberately obfuscating my point, favoring terseness over clarity, carefully avoiding counter-arguments, and other behaviors that I’m likely not even aware of. This is in addition to all the built-in biases that humans come pre-installed with to be ensured of their own correctness. This blog is one long struggle (flecked with the occasional victory) against these tendencies.

These habits are somewhat soothed by the sprinkling of caveats I try to include in whatever I write. Lots of “seems to be”, lots of “appears”, lots of “as far as I knows” and “to the best of my knowledges.” But these do an inadequate job of masking the certainty that by all rights shouldn’t be there in the first place.

The most frustrating part is that it’s not hard to do this once you focus on it. Sure, I may not be able to trace the exact path a belief takes through my cortex, but I’m reflective enough to know how much I know about something. But I have to actively consider it, and ignore my desire to be a constant source of pithiness, insight, and wisdom. And in general, I don’t, save for rare moments of lucidity (this one).

Since this isn’t good enough for me, let these few paragraphs signal my commitment to stating clearly and concisely what it is I know – no more and no less.

Complex natural phenomena typically have a size window that the laws of physics allow them to operate in, where the processes required for them to occur dominate.

Cyclones are caused by a variety of atmospheric phenomena that result in stable areas of rotating fluid. The smallest cyclones are tornados and waterspouts, which exist on a scale of 2 km (if you include the atmospheric formations that generate them). The largest known cyclone (technically, an anticyclone) is Jupiter’s Great Red Spot, maxxing out at 40,000 km in diameter. 4 orders of magnitude seperate them.

The same goes for galaxies. The smallest galaxy yet found is Leo T, a dwarf galaxy only 600 light-years in diameter. On the other end of the spectrum is IC1101, a cD type galaxy over 6 million light years across – 4 orders of magnitude larger.

Stars, luminous balls of plasma powered by nuclear fusion, also have scale limits. The lower end is around 75 Jupiter-masses, the minimum amount needed to generate nuclear fusion. OGLE-TR-122b, the smallest yet discovered, is around 100 jupiter masses, 166,800 km in diameter. The largest is VY Canis Majoris, at 3,614,000,000 km in diameter. Once again, the difference is 4 orders of magnitude.

Replicators, on the other hand, stubbornly push the boundaries imposed by the laws of physics. The smallest organism is the Mycoplasma genitalium bacteria, at a mere 200 nanometers in diameter, or 1/5,000,000th of a meter. The largest organism is the Giant Sequoia, which can grow nearly 100 meters in height. 8 orders of magnitude seperate the largest and the smallest, an enormous window to operate in. (You can in fact go larger or smaller by pushing the boundaries of what you classify as an ‘organism’.)

There are, of course, enormous selection effects at work here – it’s much easier to measure phenomena on Earth than phenomena anywhere else in the universe, so it will naturally be catalogued much more thoroughly. Nevertheless, the basic machinery of life (ATP energy source, DNA encoding for proteins, cellular structure) has adapted itself to an enormous space of possibility, one larger than almost* any other complex phenomena yet discovered.

*The only exception I can think of is black holes, which can grow without limit due to gravitational effects dominating at large scales.

Every so often, I’ll come across people that seem to have it all figured out. People that live the kind of life I want to live or act the way I want to act or think the way I want to think. The internet has made this especially easy – everything they display or put forward can be carefully sculpted and trimmed to be maximally appealing while at the same time seem to be off-the-cuff. This appearance of casual brilliance makes it easy for me to fall into the trap of idolizing said person. This, of course, is dangerous – idolization is a kissing cousin of certainty, and when you idolize someone you lap up whatever they say without pausing to think too much about it. You then proceed to shout whatever you’ve heard from the rooftops, without applying any appropriate skepticism to their words. The whole thing is one epic failure of rationality, one that I’ve committed time and time again.

It seems to happen (at least to me) on a smaller scale whenever I read a sufficiently good article. “Here’s someone who’s got it figured out” I think, about something I’m almost certainly not qualified to judge the quality of. I can only imagine how many false facts are lurking in my brain on account of a particularly persuasive writer.

But the whole REASON that idolization takes place in the first place (at least, once you get past high school and ‘has a motorcycle’ no longer serves as a suitable reason) is that they seem to have some sort of understanding that I lack. If they didn’t at least appear to have insights or perspective or SOMETHING to offer, fawning wouldn’t have started in the first place. So there’s a balance to be struck, to somehow avoid idolizing and calling them on their bullshit when it inevitably appears, but not restricting yourself from gleaning the insights they likely have to offer.

When people talk about their best teachers or their favorite classes, one story I’ve heard brought up more than once is the teacher who would say “One thing I’ll say in this lecture is wrong – it’s your job to find the mistake.” The students inevitably scrutinize each statement carefully, rigorously evaluating it’s truth (in some versions, the teacher cruelly ends the semester with a lecture that contains no mistakes, without telling his students). But the end result is always a comprehensive understanding of the subject.

It seems like a good generally applicable heuristic (always assume an error), but one has to be more deliberate in applying it to those we already respect (we tend to be skeptical of people outside our chain of respect by default). The likely default is to tend towards mental laziness and assume whatever they’re saying is correct without unpacking each claim and inference. The sort of looming problem that I can see with this is such examination is likely mentally exhausting – thinking is hard work. I’m not sure if this is the sort of thing you can eventually learn to do on autopilot, or if it’s something that you’ll constantly have to consider deliberately.

Beliefs come in two flavors: preferences, and methods of satisfying those preferences.

Preferences are completely arational. There’s no rigorous framework to decide what they should be – they’re basic axioms that have to be assumed (or, in this case, specified). It’s impossible to point to a person’s (or a dogs, or a computers) preference and call it ‘wrong’. The best you can do is show that it doesn’t fit coherently with other preferences, or runs counter to some lower-level preference. However, because humans are cut from the same cloth, people tend to share a similar set of preferences (though they may be weighted differently). Arguments about preferences tend to be of the unresolvable type.

Methods of satisfying those preferences, on the other hand, are 100% rational. There’s a correct way, an optimum that can be achieved -satisfying preferences is a math problem. Humans (and their preferences) are instantiated in the world as a big glob of atoms. Satisfying preferences means arranging these atoms in a particular way – there is one group that is superior to all other groups of atoms. (In actuality, huge swaths of groupings will be functionally identical, but we’ll ignore this). Because the behavior of atoms follow the laws of physics, some series of actions will result in that arrangement of atoms, and some will not. Determining what series of actions this is can, in principle, be calculated.

Of course actually calculating it is impossible for several different reasons, and any method of optimization actually used will abstract away a great deal of information. But because we’re in a universe based on laws of physics, at some point it HAS to become a math problem – go down far enough and there’s a set of rules.

People argue about preferences because they expect everyone to have the same ones they do. And they argue about methods of achieving those preferences because, lacking a Prime Intellect, it’s not always obvious what the best method to use is. Most arguments are a subtle yet rich blend of these two types.

According to standard economic theory, a minimum wage is inefficient and creates unemployment – people who would be willing to work for less money are instead excluded from the marketplace. Because economics is complicated and we generally don’t run experiments, it’s hard to tell if this actually happens or not (and people with various agendas to push argue fervently on both sides) – but that’s what falls out of the most basic level of the theory.

However, much of the disagreement (or at least, the initial reaction) isn’t concerned with whether it’s actually efficient or inefficient, but that it’s exploitative to pay someone less than a certain amount. We have a built-in fairness detector that the minimum wage sets off. In this sense, the minimum wage is like a real-life version of the ultimatum game. In the ultimatum game, there are two players. One is given an amount of money, and has to decide how much to offer the other player. The other player can either accept the offer (in which case the money is divided and the game ends) or reject the offer (in which case neither player receives any money).

The ‘rational’ thing to do is for the second player to accept whatever offer the first player gives him – any amount of money is better than zero. But this isn’t what happens. Offers less than a certain amount (it varies depending on the exact structure of the experiment, but is I believe around 40% of the total) are rejected, leaving both players empty handed. Our sense of fairness is powerful enough to overcome our rational self-interest.

It’s this inbuilt sense of fairness that seems to create our intuition about a minimum wage being necessary. Of course, it could very well be that a minimum wage actually is more efficient for various reasons – wikipedia lists a variety of arguments for both sides. But because the simple principles of supply and demand imply that it is inefficient, I suspect that one would not have been implemented without our built in fairness sensor.

If an insight doesn’t feel like an insight, how do you recognize it? Does it just slip by?

Humans don’t do anything without motivation. Energy is expensive, and acquiring it is risky – without the proper motivation, we default to doing nothing at all. Fortunately, the brain has access to a wide variety of chemical motivators that can be released to get us off our asses should the need arise, the chief one being dopamine. It plays a critical part in our pleasure and reward systems – it’s the carrot our brain dangles in front of us to get us to pursue a wide variety of behaviors, including learning. As such, the brain rewards useful thinking – generating an insight, such as “lions can’t climb trees” or “this sharp stick can be used as a weapon” sends a burst of dopamine that we interpret as a “feeling of knowing.” It’s this feeling of knowing that lets us know we’re on the right track.

However, it’s possible to decouple this feeling of knowing from the actual thought or insight its paired with. Consider the tip-of-the-tongue phenomena – you know that you know something, but you just can’t articulate it for some reason. You have a strong feeling of knowing without the thought it’s supposed to be coupled with. And strong feelings of knowing can be generated by a sufficiently powerful magnetic field applied to the right area of the brain. Not to mention a centuries worth of reports on religious experiences had under the influence of drugs.

As such, it stands to reason that the reverse is possible – that you can have an insight without being the accompanying feeling of knowing. In fact, it probably happens all the time – the mental connections in the brain are constantly adjusting, strengthening and weakening. It’s probably only the strongest ones that gain enough salience to enter working memory. And it’s only the strongest of those that give you that moment of clarity, that make you sit up and say ‘Yes!’, that send you running for a piece of paper to write it down before it slips away.

But without that feeling of knowing, that little burst of dopamine, it could pass right by without so much as a nod.

My guess (without any real information to back it up) is that it happens quite a bit. Maybe not total silence from the motivational system, but the equivalent of one lone cheer instead of a roaring crowd. There’s probably thousands of extremely intelligent people working boring jobs as accountants or tax collectors or insurance agents because they aren’t properly rewarded for their own brilliance.