Opinions and Ideas for Reading Research

A Supplement to Coach Your Brains Out

I recently listened to an episode of the Coach Your Brains Out podcast about reading research. I wholeheartedly agree with the ideas presented in the episode and I encourage you to listen to it, if you haven’t already. My thoughts below are responses to and tangents from the discussion. My thoughts will make more sense if you listen. Here’s a link:

One of the biggest takeaways from Andy, John, and Kasey’s discussion is that getting into research literature can be intimidating and challenging. Both Andy and Kasey point out that research articles are typically written for an academic audience. I agree with them and I, too, wish it was written in a more-accessible way. I have two ideas around the way research is written.

First, despite all our best wishes, I don’t expect academic language and article format to change to make it more accessible to non-academics. If it’s not going to change, then it’s up to you to get used to it. My favorite thing to do to get accustomed to academic-speak is to get accustomed to how just one academic (or a small group of them) speaks. If there’s a topic you’re interested in, find multiple papers in the area by the same author(s). The goal is not to learn all their arcane terms but to get used to how they talk about their work. When I first start reading papers in a new area for me, I have to remind myself that parts of the papers will be confusing until I get some lay of the land. After reading a few papers, I may reread the first one again just to see what makes more sense. Picking up on some of the specialized terms will happen but that’s much easier if you’re familiar with how some authors write first.

Second, not everything is written for you. It’s not that you shouldn’t read research articles, but you should remember that most academic journals aren’t trying to inform the general public. Scholarly articles don’t belong in newspapers because that’s not how they’re designed. I want to read academic work because academics are out there, trying to figure new stuff out. But it can be a pretty big gap between out there and what you are currently doing. If you start reading an article expecting it to fit into your current world, you’re often setting yourself up for confusion and regret. To mitigate this, I recommend reading articles looking for ideas rather than answers.

When researchers are out there, they are often working in very small niches so it’s difficult to find work that is clearly applicable to whatever particular situation you have in mind. Reading journal articles means being ready to extrapolate and try stuff without an expectation of success. It can be said that parts of research articles are written as how-to’s, meaning that another researcher should be able to read the methods and analysis sections and know, in principle how to recreate what the original researcher did. But journal articles, because they’re written from somewhere out there, aren’t how-to’s for you and me. So don’t read them like they’re supposed to tell you what to do. Instead, read them to get ideas about what to try and see what happens in your own little corner of the world. When you try an idea inspired by some research you read, you should recognize that you’re almost certainly applying an intentionally highly constrained concept to a different scenario. That’s why the internet has the concept of YMMV.

Looking for ideas in research instead of looking for answers feels really weird for most people because most of your exposure to science has been set up so you could expect certain outcomes. Think of some of the cookbook recipe-following science “experiments” you did in school. They were designed to illustrate a particular concept you had just learned about in a lecture. But those weren’t true experiments, nor were they research. In the research you read about in journals, the experimenters are very often hypothesis testing, which is very different from most science you were exposed to. To effectively read research, you have to let go of the expectation that you will find answers in the conclusion. To build on something Andy said in the podcast, the strongest language a researcher will use is that their findings suggest something. They’re being very deliberate about not presenting answers, so take them at their word and don’t treat their findings as anything more solid than they do.

Another important consideration is how you choose which papers to read at all. Often, you’ll hear about a particular paper or concept. You’ll search the internet and find something like the example pictured above. Kasey and John commented on the downfalls of only reading the abstract to draw conclusions about the paper and I agree with them. I also think the abstract is worth reading because it can function as an important sorting tool. Even though journal article titles can be very descriptive, reading the abstract can help you figure out if the paper is interesting enough to merit further reading. Think of the abstract as a movie trailer, checking it out helps you decide if you should invest the time in watching the entire movie. But don’t think that watching a trailer means you’ve watched the movie. The abstract should give you enough insight into the questions the researchers were interested in and how they went about shedding light on those questions. If the abstract doesn’t sound like it applies to whatever you’re studying, then move on. There’s plenty of other papers out there that will apply, so spend a little bit more time reading abstracts to save you from spending a lot of time reading papers that aren’t helpful.

Lastly, maybe you’re familiar with using search engines to find papers, maybe you’re even used to using Google Scholar. (If you don’t use Google Scholar, I highly recommend it.) Connected Papers is another tool that I really like for finding more authors and papers after I find a topic or author I want to dig further into.

With all that said, I think it’s important to say that you don’t have to read all the research yourself. Sometimes, researchers get around to writing their own books about their research. In the podcast, Kasey mentioned Carol Dweck’s work on mind set as an example of work that has become popularized at the cost of losing much of the nuance in the original work. Instead of reading all the hot takes about her work, go read her own book on the subject. What about expertise and deliberate practice? Instead of reading Outliers, read Ericsson’s own book about his work. Want to learn about heuristics and biases? Don’t read The Undoing Project, read Kahneman in his own words in Thinking, Fast and Slow instead. What about ecological dynamics? Rob Gray’s books are great alternatives. While those books may be more challenging to read than popular science versions, they maintain much more of the original context than popular authors do while being much more readable than journal articles.

Another way to get your fill of research without reading the articles directly is to find reliable science communicators. Malcolm Gladwell, for all his writing skill (as Kasey pointed out), isn’t very reliable when it comes to science. Neither is Michael Lewis, despite Moneyball and The Undoing Project being such amazing books. But there are others, like Steve Magness and Brad Stulberg, who synthesize lots of scientific research after reading lots of articles for you. They are great at giving you an appropriate amount of nuance, context, and qualifying language. I think the key thing to look for in reliable science communicators is the ability to translate academic speak well. Avoid writers with a tendency to impose their own meaning on someone else’s work.

Happy reading and learning!

---