Epistemic Status: I thought about and wrote each paragraph in 10 minutes total, with slight editing afterwards.

I hope I'm not too late to the party! 

I wrote this up quite a few months ago and found that I delayed indefinitely editing it before publication. I decided it's probably best to post a not-maximally-edited version of my final exam. 

This is my first post on LessWrong, and any remarks, notes, requests for revisions,  and so on will be welcome!

Cognitive defect: calendar view leaky abstraction

When using a calendar or task management app, one can often judge quickly the length of a task by it's physical size on the screen. This is a useful feature of the calendar view. 
This heuristic is however wildly wrong for small and short tasks. Since the tasks need to be readable, the app often displays them as taking somewhere between 5 to 15 minutes, depending on the font size and so on.
Having a large set of small 2-10 minutes task on the screen may thus feel (incorrectly) overwhelming. 
The size of a task on the screen is a leaky abstraction (of it's length in time).

Rationality skill: course correction

The rational skill I'd like to talk about is course correction.
Course correction helps install replacement habits to unoptimized habits.
For example, consider the habit of not using the correct fingers when touch typing.
For that habit, a course correction might look like that: notice that you are not using the correct finger, remember that you'd like to install the habit of using the correct finger when touch typing, delete the word or letter and (while being aware to perform correctly) redo the action.
This will be useful in two regards: 
1. You will actively perform the desired action many times, getting used to it
2. From the desire to optimize, your brain, not wanting to waste time on doing and undoing and action will get used to only performing the correct version.

Before you may apply course correction as a tool you need to first strongly identify with the desire to optimize. Further you need to understand how to identify the mistaken habit in real time. I find that it usually help to identify ahead of time what sort of activities may lead to that sort of mistake. 
 

For the example of touch typing, you need to be aware of 
1. Which is the correct finger for each letter, and to have an instinctual feeling of an uncomfortable hand movement when typing which corresponds to a letter being typed with the wrong finger.
2. That when you are starting to write you enter an environment where this mistake may pop up.
You need to think about both of these ahead of time, when setting up the course correction.
After that it might be good to put a 5 minute YODA timer for practicing the course correction in one (or many) of the environments you picked: making the mistake on purpose, focusing on how it feels to make it, and tying it to the corrected habit.

Rationality principle: Rediscovering "don't jump to an answer"
 

Posthoc: I didn't have too good of an idea for an original rationality principle. While trying to write out my general thoughts on the question of "What makes some approaches for solving a problem better than others?" I ended up rediscovering Hold Off On Proposing Solutions. Following is pretty much my thought stream at the time, edited after the fact. It's ended up quite meta.

What differentiates effective and productive directions of progress from unproductive ones? How might one find out if a given direction is productive? What is a direction of progress, even? 

A direction of progress is essentially an idea which attempts to get you closer to your goals either directly, or via leading you to a closer point.
For example, me trying to think of things resembling a rationality principle when trying to come up with a topic for this essay wasn't a good direction. It wasn't a good direction because it asked me to do the whole thing at once, and it offered no starting step to solve the problem. It also made me tense and made me feel lost. 

So what does a good direction look like? My 5 minutes are up but I'll continue writing (and thinking) anyway. 

I think that the framework I want to propose is as follows - good (or, at least, non-horrible) direction are those who by following them you learn something other then just directly the answer.
When the direction is just "know the answer", that is not actionable.
To know the answer will thus require you to learn things adjacent to your main goal which may contribute to the answer. 
In particular, looking for direct examples of adjacent things will often be a good direction, as is trying to understand the meanings of words, or looking for direct other examples, playing with toy examples, so on and so on.

The principal thus is: When aiming to answer a question, ask further questions about adjacent ideas. You may perform this recursively until something yields an answerable question, and from there you may continue with other methods.
This is similar to "don't jump to an answer" idea, but it's my take and it was hard to write which is cool! :)

Thanks for reading!