How to actually apply the things you learn

The problem: 
If you don’t apply what you learn then learning loses most of its value. For example, suppose you learn that exercising makes you happier and increases your lifespan, both goals you greatly desire. If you don't exercise as a result of learning that, then it didn't do much good to you. Indeed, it might have made you feel stressed out, which in turn decreases your happiness and lifespan! 

This is not to say that unapplied knowledge is always useless. For example, some people get intrinsic enjoyment out of learning. There is also the potential benefit of “working out” your mind and increasing its ability to learn things quickly. 

On the whole, however, a huge proportion of the value of learning comes from applying what you've learned. Yet for some reason people are generally not great at that. 

How it applies to doing good
There are many things you can learn about doing good in the world that if they aren’t applied end up being not very useful. For example:

• Skepticism
  
• The scientific method
  
• Measuring the right things
  
• Base rates

The list goes on and on. Indeed, if what you are learning from these blog posts is not contributing to a change in how you act or think, then reading them is being of little benefit to you. Except for your enjoyment of reading them. 

Some solutions
Fortunately there are solutions. Here are some that have worked for us in the past:


1. Rubrics and checklists. 

Evidence: Rubrics and checklists have been catching on in the rationality community, and for good reasons. Studies have found that having checklists improves doctor and nurse performance. Rubrics are like checklists with the added nuance of evaluating yourself with a number or qualitative assessments on each item. Think the grading criteria you had on your school assignments except you get to grade yourself. This has been found in meta-analysis of educational psychology to be the highest impact way of improving school performance. Particularly when you “grade” yourself and then set goals on how to improve. 

Implementation Examples: We used a rubric for grant-writing where we wrote down common aspects of a persuasive and good application (such as catchy title, appealing to readers’ motivations, etc). After writing the first draft of the application we would go through and grade ourselves on different aspects, then go back and improve the weak areas. Some other examples are a rubric for qualities of a good study or a good charity. Also, for those of you who subscribe to the motto, “the meta the bettah”, here’s a checklist on how to make a good checklist. 


2. Explaining it to others

Evidence: Studies have found the best way to retain information and to improve understanding of materials is to explain it to others. Additionally people are more likely to follow through on things that are publicly declared which happens every time you explain an idea to somebody.

Implementation Examples: This blog itself (meta!). Writing this blog will help me understand applying ideas better because I’m explaining it to you and I am making a public declaration of my intention to use these strategies. So if you ask me and I am not doing these things I will be insanely embarrassed and have to disappear into the Canadian wilderness to avoid your judgment. Thus writing blogs are a good way to explain things to others. Another way is simply to discuss all of the ideas you write about with friends in person or online. 


3. Spaced repetition. 

Evidence: While remembering is not a guarantee of application it does increase the chances. If you do not remember the results of a study you read a year ago, you cannot use it when the time comes where it is relevant. Spaced repetition has been found by psychologists to be one of the best way to remember things long term.

Implementation Examples: Download mnemosyne or other spaced repetition software and input the ideas you’d like to implement. 


4. Learning with an application in mind

Evidence: For example, accountants do better years later if they are taught by going through a pretend accounting case. It has also worked for me quite a lot in the past as anecdotal evidence.

Implementation Examples: Currently I am learning more about research methods and statistics. To make sure that it gets fully absorbed and applied I have two studies I’m applying them to - the leafleting study we are running and the RCT GiveDirectly ran on its impact. This way, after every new feature of a good study I learn about, I can ask myself “Did GiveDirectly do this? Did I do this in my study? If not, how could I have made it better?”. 

This is definitely not an exhaustive list and if you have any other ways to ensure application of ideas, please add them below in the comments.