I am intimately aware of the errors in my thoughts and the sins of my soul. I can hear the Type-A asshole screaming like a stolen mind in the Sunken Place when I am in my fifth hour of streaming a Netflix TV show I don't care about. A plethora of self-help authors' voices ring out when I spend another day not implementing any of their advice at all.
The CFAR Participant Handbook, amazingly, does not me feel shitty for being a human being. The genre imperative, "I need to become productive achieve my dreams" is reconfigured as "Wait what are my dreams?"
Note: CFAR, the Center for Applied Rationality, is a consultant group that goes to Silicon Valley corps and organizations or otherwise accepts students, and teaches a bunch of "rationality" techniques. These are day/weeklong classes that I, being an East Coaster, have had to consume via blog posts and regurgitated efforts of graduates from the program. This actual participant handbook, released in 2020, was new to me in its _systemization_ of the random things I have been reading.
The handbook itself is a list of techniques, meant to be used by students after their classes as a reference guide. I have found it to be a useful without ever attending the classes at all.
As a handbook, they never really lay out an explicit framework for how the different techniques work together. They're presented as a rather separate concepts that solve specific genres of problems. Goal factoring, for example, solves cases when you have an action whose end state might be achievable by other means. The Hamming Technique is used for figuring out the importance and prioritization of what you're working on. These two may seem different, but I think ultimately most of these techniques come to specialized answers to three broad questions:
1. How do I program my Pet Human to be the happiest it can be? 2. How do I really talk to and engage with what my feelings of happiness would be? 3. How do I improve my ability to think about Problems?
There is a last category of question, maybe described as "Why the fuck aren't you doing this?"
CFAR really buys into the Kahneman view of the human brain, which is the split between System 1 and System 2. The former, famously, is quick and general and cheap, while the latter is slow, thoughtful, and energy depleting. I prefer a similar but different main model: the Pet Human and the Human. The Pet Human, is like my cats, incredibly automatic and babylike. It needs to have a schedule and its hand held and it goes crazy over treats.
One goal of CFAR's techniques is to figure out how to make the Pet Human happy. Concepts like "Trigger, Action, Planning" and "Taste & Shaping" are ways to train the Pet Human to be more aligned with its higher goals. Indeed, these techniques might be described as "meta" classical and "meta" operant conditioning, because the conditioning happens to a large degree in the mind.
However, before you can even begin programming the Pet Human, you have to know what it wants. That is where the real heart of the CFAR techniques comes. "Goal Factoring", "Aversion Factoring" and "Focusing" are all different ways of saying, "Hey, Pet Human, what is it you actually want?" It is similar to me when I'm trying to figure out why my cat is screaming from the top of the couch. I have to watch her, check her food and litter and water, and intentionally see how she is behaving.
The fact is, while we probably do intentionally observe and interrogate our cats, we do not intentionally interrogate ourselves.
Once you know what your Pet Human wants, you then need to actually go about solving their problems. Formulating and solving them requires concepts like the "Unit of Exchange" and the "Area Under the Curve". Basically, you need techniques that fight against innate human errors that we are prone to make. You can also externalize the decision process through "Systemization" or internalize improvement through "Deliberate Performance".
The optional question, "Why the fuck aren't you doing this?" is pretty straightforward: lots of CFAR advice comes down to you're not solving things because your Pet Human is in a feedback loop of procrastination, confusion, and guilt. Break that out by _just doing it_. Literally! just Shia LaBeouf it. "Resolve Cycles" are periods of time that you simply solve a problem without planning on it. It's how I solved a tax dispute, and, by golly, it worked despite me procrastinating on it for literally half a year. A generalized problem version is the "Eat Dirt" technique, where you just do things semi-intentionally, until you feel out what you're actually intending to do.
The answer to these four questions, which are described in 13-35 concepts, creates a useful framework to approaching a wide range of problems. What I found interesting is that many of these techniques are not new; they're systematized versions of things that we naturally do. By applying them coherently and together, they should amplify our natural self-help tendencies.