Instrumental Rationality 1: Starting Advice

[Instrumental Rationality Sequence 1/7. Repost from LW]

[This section goes over 4 concepts that I think are important to keep in mind before we start the other stuff. We go over caring about the obvious, looking for ways to apply advice in the real world, practicing well, and holding realistic expectations.]

In Defense of the Obvious:

[As advertised.]

A lot of the things I’m going to go over in this sequence are sometimes going to sound obvious, boring, redundant, or downright tautological. I’m here to convince you that you should try to listen to such advice anyway, even if it sounds stupidly obvious.

First off, our brains don’t always see all the connections at once. Thus, even if some given advice is apparently obvious, you still might be learning new things.

For example, say I told you, “If you want to exercise more, then you should probably exercise more. Once you do that, you’ll become the type of person who exercises more, and then you’ll likely exercise more.”

The above advice might sound pretty silly, but keep in mind that our mental categories for “exercise” and “personal identity” might be in different places. Sure, it’s tautologically true that someone who exercises becomes a person who exercises more. But if you’re not explicitly thinking in terms of how your actions change who you are, which is what the tautology is pointing at, then you’ve still learned something new.

Humans are often weirdly inconsistent with our mental buckets—things that logically seem like they “should” be lumped together by logical implication often aren’t.

By paying attention to even tautological advice like this, you’re able to form new connections in your brain and link new mental categories together, perhaps discovering new insights that you “already knew”.

Secondly, obvious advice is often used as a label for what everyone know works. If your brain is pattern-matching something as “boring advice” or “obvious”, you’ve likely heard it before many times before.

For example, you can probably guess the top 5 things on any “How to be Productive” list—make a schedule, remove distractions, take breaks, etc. etc. You can almost feel your brain roll its metaphorical eyes at such dreary, well-worn advice.

But if you’ve heard these things repeated many times before, this is also good reason to suspect that, at least for a lot of people, it actually works. Meaning that if you aren’t taking such advice already, you can probably get a boost by doing so.

If you just did those top 5 things, you’d probably already be quite the productive person.

The trick, then, is actually doing them.

Lastly, it can be easy to discount obvious advice when you’ve seen too much of it. When you’re bombarded with boring-seeming advice from all angles, it’s easy to become desensitized.

What I mean is that it’s possible to dismiss obvious advice outright because it sounds way too simple. “This can’t possibly work,” your brain might say, “the secret to getting things done must be more complex than that!”

In philosophy, there’s this idea of a “hedonic treadmill”, an idea based off our human inclination to compare experiences relatively. For example, if you have something tasty, like a chocolate bar, then you’ll need something even tastier the next time, like a tiramisu cake. A more stereotypical example might be a drug user seeking out an even stronger high for their second experience.

The point is that, as you are exposed to more and more pleasures, you find yourself on a treadmill climb to seek out ever more delicious experiences (because, by comparison, everything else will seem dull).

There’s something akin to the hedonic treadmill happening here where, after having been exposed to all the “normal” advice, you start to seek out deeper and deeper ideas in search of some sort of mental high. What happens is that you become a kind of self-help junkie.

As a self-help junkie, you end up adopting quite the contrarian stance—you reject the typical idea of advice on grounds of its obviousness alone. There’s a certain aesthetic to being the cool kid who knows that simple advice isn’t enough to solve their complex, multi-faceted problems.

You can end up craving the bleeding edge of crazy ideas because literally nothing else seems worthwhile. You might end up dismissing obvious helpful ideas simply because they’re not paradigm-crushing, mind-blowing, or mentally stimulating enough.

If this describes, might I tempt you with the meta-contrarian point of view?

Here’s this for a crazy idea: One of the secrets to winning at life is looking at obvious advice, acknowledging that it’s obvious, and then doing it anyway.

(That’s right, you can join the even cooler group of kids who scoff at those who scoff at the obvious!)

You can both say, “Hey, this is pretty simple stuff I’ve heard a thousand times before,” as well as say, “Hey, this is pretty useful stuff I should shut up and do anyway even if it sounds simple because I’m smart and I recognize the value here.”

At some point, being more sophisticated than the sophisticates means being able the grasp the idea that not all things have to be hyper complex. Oftentimes, the trick to getting something done is simply to get started.

Because some things in life really are obvious.

++++

Hunting for Practicality:

[This is about looking for ways to have any advice you read be actually useful, by having it apply to the real world. ]

Imagine someone trying to explain exactly what the mitochondria does in the cell, and contrast that to someone trying to score points in a game of basketball.

Someone could take classes to learn how to get better at each of those two things.

Yet, there’s something clearly different about what each person is trying to do, even if we lumped both under the label of “learning”.

It turns out there are roughly two types of knowledge you can learn: declarative and procedural knowledge.

Declarative knowledge is like the student trying to puzzle out the mitochondria question; it’s about what you know. It’s about how your concepts link to one another, like how you can know that Paris is the capital city of France, or that it takes about 20 minutes to walk a mile.

In contrast, procedural knowledge, like the fledgling basketball player, is about what you do. It’s about how you actually carry out certain actions, like how you learn to throw a frisbee well, or how to ride a bike.

I bring up this divide because many of the techniques in instrumental rationality will feel like declarative knowledge, but they’ll actually be much more procedural in nature.

For example, say you’re reading an essay on motivation, and you read about how “Motivation = Energy to do the thing + a Reminder to do the thing + Time to do the thing = E+R+T”.

What’ll likely happen is that your brain will form a new set of mental nodes that connects “motivation” to “E+R+T”. This would be great if I ended up quizzing you “What does motivation equal?” whereupon you’d correctly answer “E+R+T”.

But that’s not the point here!

The point is to have the equation actually cash out into the real world and positively affect your actions. If information isn’t changing you view or act, then you’re probably not extracting all the value you can.

What that often means is figuring out the answer to this question:

“How do I see myself taking different actions as a result of having learned this information?”

With that in mind, maybe you generate some examples and make a list in response to the question.

Your list of real-world actions might end up looking like:

  1. Remembering to stay hydrated more often (Energy)

  2. Using more Post-It notes as memos (Reminder)

  3. Start using Google Calendar to block out chunks of time (Time).

The point is to be always on the lookout for ways to see how you can use what you’re learning to inform your actions. Learning about all these things is only useful if you can find ways to apply them.

As we go forward, I’ll try to give concrete and useful examples for all the ideas we go over, but you want to find ways to move past the simple pattern-matching. You want to do more than have empty boxes that link concepts together. It’s important to have those boxes linked up to ways you can do better in the real world.

You want to actually put in some effort trying to answer question of practicality.

Knowledge might be power, but you also often need to act on it.

++++

Actually Practicing:

[This is about knowing the nuances of little steps behind any sort of self-improvement skill you learn, and how those little steps are important when learning the whole.]

So on one level, using knowledge from instrumental rationality is about how you take declarative-seeming information and find ways to utilize it in real world actions.

That’s definitely important.

Another thing, though, is to note that the very skill of “Generating Examples”—the thing you did in the above essay to even figure out which actions can fit in the above equation to fill in the blanks of E, R, and T—is itself a mental habit that requires procedural knowledge.

What I mean is that there’s a subtler thing that’s happening inside your head when you try to come up with examples—your brain is doing something—and this “something” is important.

It’s important, I claim, because if we peer a little more deeply at what it means for your brain to generate examples, we’ll come away with a list of steps that will feel a lot like something a brain can do, a prime example of procedural knowledge.

For example, we can imagine a magician trying to learn a card trick. They go through the steps. First they need to spread the cards. Then comes the secret move. Finally comes the magical reveal of the selected card in the magician’s pocket.

Even though the audience experiences the whole trick as one magical unit, the magician knows that it’s really made up of all those little steps.

Likewise, we can apply the same analogy to things we’ll learn in instrumental rationality, to the mental habits we’ll go over.

If we spent some time really looking at the little steps of coming up with examples, we could describe it in detail. The skill of Generating Examples, with a reductionist view, might look something like this:

Technique: Generating Examples

  1. Imagine the “skeleton” of the concept you are trying to fit an example to.
    EX: Anne is trying to come up with an example of what is a System 1 process. She knows it’s about fast and sometimes mistaken thinking, so she uses that as the “frame” to search for examples.

  2. Look for things in your everyday life that fit.
    EX: Anne thinks about things in her daily routine which don’t require much thought. “Maybe brushing my teeth?” she wonders.

  3. Think about books, movies, or other pop culture if real life doesn’t prove fruitful.
    EX: Anne thinks about a movie about a character who always gets into trouble because of their quick wit and fast tongue. “Hmm, maybe also the sort of social responses we give count as S1 responses?” she thinks.

++++

The idea here is to describe any mental skill with enough granularity and detail, at the 5 second level. What I mean by that is that you should be able to describe such that you’d both be able to go through the same steps a second time and teach someone else.

That means having a very deep understanding of exactly what little steps you’re going through in your head to produce the skill.

Now of course most of us already know how to generate examples, so the above “technique” formulation might seem a little alien, as it’s already something we do without much explicit, conscious input.

However, when we move on to more novel and complex habits of mind, techniques that involve moving your brain in new ways, then having a good understanding that these steps are things you do becomes quite important.

A basketball player doesn’t strongly improve just by watching the NBA. Likewise, figuring out this instrumental rationality stuff is not a spectator sport either. You need to really go through the mental motions in your own head and act on them.

This is what I mean when I say that mental habits are procedural.

In addition to figuring out the practicality and the little pieces, it’s important to find opportunities to use these different skills.

You can’t get better at doing something unless you, y’know, actually do it.

++++

Realistic Expectations:

[An essay about having realistic expectations and looking past potentially harmful framing effects.]

There’s this tendency to get frustrated with learning mental techniques after just a few days. I think this is because people miss the declarative vs procedural distinction.

(But you hopefully won’t fall prey to it because we’ve covered the distinction now.)

Once we liken the analogy to be more like that playing a sport, it becomes much easier to see that any expectation of immediately learning a mental habit is rather silly—after all, no one expects to master tennis in just a week.

So, when it comes to trying to configure your expectations, I suggest that you try to renormalize your expectations by treating learning mental habits more like learning a sport.

Keep that as an analogy, and you’ll likely get fairly well-calibrated expectations for learning all this stuff.

Still, what, then, might be a realistic time frame for learning?

We’ll go over habits in far more detail in a later section, but a rough number for now is approximately two months. You can expect that, on average, it’ll take you about 66 days to ingrain a new habit.

(There’ll be a lot more on habits in Part 4.)

Similarly, instrumental rationality (probably) won’t make you a god.

Disappointing, I know.

Still, in my experience, studying these areas has been super useful, which is why I’m writing this at all. Your own mileage will vary depending where you are right now, but this serves as the general disclaimer to keep your expectations within the bound of reality.

Here, the main point is that, even though mental habits don’t seem like they should be more similar to playing a sport, they really are. There’s something here about how first impressions can be rather deceiving.

For example, a typical trap I sometimes fall into is missing the distinction between “theoretically possible” and “realistic”.

I end up looking at the supposed 24 hours available to me everyday and then beating myself up for not being able to harness all 24 hours to do productive work. “After all,” my brain says, “that’s all time you could be using to do things!”

But such a framing of the situation is inaccurate; things like sleep and eating are often very essential to maximizing productivity for the rest of the hours! Just because it looks like I could “in theory” get additional work time, self-care is also an important factor that we easily miss!

So when diving in and practicing, try to look a little deeper when setting your expectations. Bias towards pessimism. No one likes to hear it, but the chances of you actually turning your life around at any given moment are likely slim.

That’s just how things work. Disappointing, for sure, but that’s all the more reason to be suspicious if you’ve got too rosy expectations of how things will turn out.

First glances tend to be deceiving.

++++

Next essay

Comments

sorted by
magical algorithm
Highlighting new comments since Today at 10:39 PM
Select new highlight date
All comments loaded

I hadn't read these sequences as part of LW 1.0, so thank you very much for bringing them back into the spotlight. Is there contained within them a listing of habits that have been useful to those aspiring to implement instrumental rationality? Is there a compendium of what obvious advice is on offer in various domains?

In later sections, we'll definitely get into *a lot* more concrete examples for things like planning and habits. (I made sure this was a priority when writing the whole sequence ^_^).

In the meantime, for some good compendiums, LW 1.0 has the Boring Advice Repository v1 and v2 that you might like:

http://lesswrong.com/lw/gx5/boring_advice_repository/

http://lesswrong.com/lw/lih/2015_repository_reruns_boring_advice_repository/

And also, off the top of my head, here are some things which I think are often quite useful:

  1. If you want to remember something, write it down. (e.g get a notebook for thoughts / to-dos.)

  2. If you need to get something done, set aside a chunk of time to do it. (e.g do some light scheduling.)

  3. Setting a 5 minute timer to think / go meta about you're going to do is often very useful.

  4. Journal your thoughts to keep track of stray thoughts and how your thinking changes.

  5. Whenever you learn about something abstract, ask for concrete examples.

This is a really thoughtful and interesting essay and I'm looking forward to the next thing you post!

Thank you for the positive feedback + reinforcement! It definitely helps make the whole process more rewarding.

I'll be slowly trying to edit and port the instrumental rationality sequence (including Planning 101 and Habits 101) over, seeing as the LW 2.0 revival is in effect. The current goal is one essay a day.

This is very practical! I'm looking forward to the next posts in this series; I hadn't read it yet on LW 1.0.