How I Research

I frequently do science review articles, e.g., on burnout and MDMA.  A friend recently asked for a quick start guide to doing his own, which I am sharing now in the spirit of “done is better than perfect”. I use examples from two recent projects:

  • my report on the problems and benefits of distributed work teams (which may or may not be publicly released- do comment if you think it would be useful to you).
  • my friend’s report on different sleep interventions, specifically trying to figure out which of the things he thought ought to be tested already had been tested.



  1. Search peer reviewed articles.
    There’s a tendency to treat peer review as kind of a totem, and that’s incorrect- lots of bad work gets past peer review and lots of perfectly good work is done without it. For these particular examples, there’s so much crap running around on non-peer-reviewed internet that this was necessary to limit the scope, even though you’ll miss quality content like “that supplement that worked for me for a few months
  2. The google scholar chrome plugin. If you find the abstract of a peer-reviewed paper but need the full text, this  plugin will sometimes find it for you.
  3. Sci-Hub. This keeps getting taken down because it’s legally dubious, but if google scholar can’t find you the full text of a paper, sometimes this can.
  4. Libgen: Sci-Hub for books.
  5. If you can’t find a paper any other way, e-mail the author explaining why it’s useful to you and ask for a copy. People like hearing that their work is being used.


Things to Try

  1. Google scholar search “[thing I care about] + meta-analysis” or “[thing I care about] + review”
  2. Google scholar the obvious thing (“sleep”, “distributed team”).
  3. Ask regular google for exactly what you want. E.g., scholar searching for “sleep intervention meta-analysis” was getting me a bunch of papers on sleep as an intervention for other problems.  I regular-googled “list of sleep interventions with citations” and found a promising paper.
  4. Once I find a promising paper:
    1. see if it uses different terms of art than I have been using, and retry step one. E.g., for sleep research I started with “sleep” when I should have used “insomnia”, and for the business research I thought of “remote teams” and “distributed teams”  but also needed to check “virtual teams.”
    2. Look at papers the paper cited (especially if it is a review paper). Reading the paper will sometimes draw your attention to papers you wouldn’t notice from their titles alone.
    3. Look for forward citations  (click “cited by” on google scholar) (especially if it’s not a review paper). This is often better than backwards citations because it gets you more recent papers.
    4. Look up other papers by the author (either on their website or by clicking their name on google scholar) and check for interesting titles.
  5. Amazon or regular-google search for books on the topic, investigate their authors, who they cite, and their keywords. This is often what I’ll do if I’m not at all familiar with an area, to get myself situated.
  6. Graduate dissertations are also great for orienting yourself, because they have do to a topic review.

Knowing I’m Being Tricked is Barely Enough

I think it was in Terry Pratchett’s Going Postal that a con man mused that the easiest people to rip off were other con men, or at least those who aspired to be so, because all you had to do was make it look like they were taking advantage of you. Honest men wouldn’t fall for it because they weren’t willing to rip you off.* It’s interesting watching myself fall for that.

A year ago Audible (owned by Amazon) offered me a year’s membership (=12 credits for free books, plus discounts on other books and some miscellaneous perks) for $100. That’s almost half off the per credit price (which itself can be can be half the price of buying a book with money, although those prices are clearly set to encourage subscribing rather than to be paid). I was about to start reading a bunch of dense history tomes, so this seemed like a pretty good deal. Then between my plan changing and the library gods being generous, I ended up not using a single credit. Suddenly 11 months had passed and Audible e-mailed me telling me they would renew my membership in a month.

“NBD, I’ll just cancel my account and keep the credits for when I need them” I thought quietly to myself. Only it turns out unused credits disappear if you cancel your membership. To keep them, I’d have to renew my subscription.  But Audible only lets you keep 18 credits at a time. If I renewed for another year, 6 credits will immediately expire, meaning I’m paying for 12 credits but only getting 6. So I went on a spree buying books off my Goodreads queue and gave a few as gifts. Actually I gave one too many gifts, because I realized later there was another good candidate I didn’t have the credits for.

Audible to the rescue. No sooner had I spent my last credit than they offered me three for $30 each- more per credit than I’d paid for my first membership, but less than the annual membership. I went as far as picking out two more books before realizing this was stupid, I had gone an entire year without buying a single audio book, I did not need three more credits.

Then my roommate told me $10/credit isn’t even that good a price, they were sure to offer me a better one when I actually cancelled. And I wanted it. A product I empirically had to be forced to use, and I felt compelled to buy more because it was cheap.

I put this down to two things- “Audible subscriber” was a nice identity to have and I enjoyed the feeling of pulling one over on Audible. I was the dishonest man letting the con man fool him. And I knew this was happening and it was still an act of will that I actually cancelled my subscription.

This story has a happy ending- well, except for the part where I paid $100 for a bunch of books I only marginally wanted. I probably still captured some surplus. But this can not possibly be the only trick Amazon is playing on me, and I don’t know what to do about it.


*I tried to find an exact quote for this and instead found “There is a saying ‘You can’t fool an honest man’ which is much quoted by people who make a profitable living by fooling honest men.” But I’m pretty sure the part I quoted was there too.

Letters to the Future

Like many of you, I have parents. They are in reasonable mental and physical health now, but either they die by physical trauma or live long enough that that is no longer true. If they follow their parents and grandparents they will spend a long period where they are alive and possibly even enjoying it, but lack the mental capacity to manage their own affairs. No one wants this to happen, but it probably will.

If/when the decline happens, it will be useful for a trusted person with more mental capacity to have power of attorney over my parents. I have one sibling, and he spent high school saying “remember, I’ll choose your nursing home” every time our parents did something he didn’t like, so it’s probably me. Unfortunately, one of the hallmarks of incompetence in general is inability to recognize incompetence, and this only gets worse when you mix it with the belligerence of old age and dementia. I’m trying to imagine a more uncomfortable conversation than convincing the person who used to change my diapers that they’ve suffered an Algernon and should let me handle their finances, and there aren’t many.

As an attempted dodge, I asked my dad to write a letter to his future self that I could give him when I felt it necessary. The eventual product included both some tests he could run to see if now was power of attorney time (such as writing out multiple checks for the same bill), and stories of his parents and grandparents who had waited too long.

I don’t have any tips for how to have this conversation with your parents, because my dad is quite reasonable about these things and I punted the conversation with my mom to him. But it is something I recommend, early enough that your parents can consider it hypothetical.

Cortisol Manager for Sleep

I wanted to do a whole rigorous review of all of the individual ingredients, but my new role at work is sucking up all my time and every time I mention “I found a supplement that cured my sleep problem” people demand to know what it is, so screw it, here’s the unscientific version.

Cortisol is the long term stress hormone. High cortisol at night can prevent you from falling asleep, or cause you to wake up in the middle of the night. I had this problem a lot, usually at 3 AM.  I started taking Cortisol Manager and started sleeping through the night, which, wow, is everything I dreamed it would be. It is not the only reason I went from working 60 hours/week to working 90, but it was a contributor.

[Note: “work” means “consciously chose to do a thing, of any kind, as opposed to looking up and realizing I’d just spent 40 minutes on Reddit.”, although my actual job-work hours did go up too]

Cortisol Manager has a number of ingredients, two of which I know anything about:

  • Ashwaganda, which is an “adaptogen”, raising or lowering cortisol as you need it. This is good, because cortisol is extremely important for getting out of bed in the morning. Taking ashwaganda alone does not seem to help me.
  • Theanine, an amino acid that lowers anxiety. Theanine alone leads to higher quality sleep for me, and helps my anxiety enough to be used for dental visits before I got a prescription for something stronger.


I’ve been on and off CM in the past. At points it depressed on cortisol too much (although right now I need two tabs/night), other times it just didn’t feel necessary. But it is extremely necessary for me right now, and at least one friend has benefited tremendously from me sharing the recommendation.


I have published a lit review on burnout. You can read it at the new EA Forum. I think a lot of the value will come from discussion in the comments and I want to keep them all in one place, so please comment there rather than there.

Epistemic Spot Check: The Dorito Effect (Mark Schatzker)

Epistemic Spot Checks is a series in which I fact check claims a book makes, to determine its trustworthiness. It is not a book review or a check on every claim the book makes, merely a spot check of what I find particularly interesting or important (or already know).

Today’s subject is The Dorito Effect, which claims that Americans are getting fat because food is simultaneously getting blander and less nutritious, and then more intensely flavored through artificial means. This is leaving people fat and yet malnourished.


Claim: Humans did not get fatter over the last 100 years due to changes in genetics.
True. People are fatter than their ancestors, indicating it’s not a change in genetics (although genetics still plays a role in an individual’s weight).

Claim: Casimir Funk discovered that an extract of brown rice could cure beriberi in chickens.

Claim: In 1932, the average farm produced 63 sacks of potatoes/acre. By the mid 1960s, it was 200 sacks/acre.


Claim: Everything is getting blander and more seasoned.
More seasoned.
Blander food.
Note that both sources were provided by the book itself.

Claim: “We eat for one reason: because we love the way food tastes. Flavor is the original craving”.
This doesn’t jive with my personal experience. I definitely crave nutrients and am satisfied by them even without tasting them.

Claim: “In 1946 and 1947, regional Chicken Of Tomorrow contests were held.”

Claim: Over time the Chicken Of Tomorrow winners consistently weighed more, with less feed and less time to maturity.

Claim: Produce is getting less nutritious over time.
True (source provided by author).


Extremely trustworthy, and therefore worrisome, given the implication that food is becoming inexorably worse. Dorito Effect is unfortunately light on solutions, so you might just freak yourself out to no purpose. On the other hand, if you’re looking for a kick to start eating better, this could easily be it.