From Abstract to Discussion: A Primer on Reading Scientific Papers

Building Scientific Literacy and Learning the Language of Science

This edition of the Unbiased Science newsletter is authored by Dr. Sarah Scheinman, neurobiologist and co-host of our podcast. Fresh from attending the Society for Neuroscience Conference, Sarah has research on the brain (see what we did there?). As someone who regularly reads and writes scientific papers, she's excited to provide this primer on how to approach scientific studies.

In an upcoming newsletter, Dr. Jess Steier will share her perspective as a data scientist with expertise in research design. She'll help identify important red flags related to the validity and reliability of research studies from a design standpoint. Stay tuned for this follow-up!

—

Today, I’m going to give you some practical advice for reading a scientific journal article. It is my personal belief that anyone, anywhere, regardless of background and education, can learn this skill. But, as with any skill, learning it requires patience and lots of practice. These kinds of articles can seem daunting at first; dense, bland, and exceedingly technical. And that’s because, well, they are! Science is a language, and if you’re not “fluent” it might seem entirely inaccessible. Fortunately, our brains are hard-wired to learn, with entire regions dedicated to just this pursuit.

But why is it so important to learn how to read scientific journal articles? You can think of them as primary source documents in the study of history. Primary sources provide firsthand accounts of historical events, offering direct evidence of what actually occurred. These sources serve as foundational materials for the construction of historical narratives, allowing for the building of arguments and interpretations based on original records and accounts. Scientific journal articles contain raw data, which is the scientific equivalent of original facts/truths. Secondary sources such as news outlets, often attempt to describe the data, but sometimes fall short because of incorrect interpretations/analysis. Unfortunately, whether intentional or not, this is how dangerous misinformation is spread. By learning to read scientific journal articles, you can gain the ability to engage directly with the data yourself, drawing your own informed conclusions and analyses rather than relying on others. This skill will empower you to navigate the dynamic landscape of scientific information with confidence and discernment.

So, let’s dive in.

When reading a scientific journal article, the most important thing is not to get bogged down in the details and to try and extract the main ideas. This is much easier said than done, but the gist is to finish the article and be able to identify the “why,” “how,” “what,” and “where” of the study. Fortunately, the structure of an article is focused around answering these exact questions. So, let’s talk about the anatomy of a journal article.

Abstract – (a “primer”)

It all begins here. The abstract is a brief (~200 words) summary of the study. Its purpose is for scientists to be able to quickly determine the relevancy of an article before deciding whether or not to devote the time to reading it. Journal article titles can sometimes be misleading and are often vague and/or convoluted. The abstract gives you an idea of what you are about to read, kind of like a book jacket summary. But just as you shouldn’t judge a book by its cover, you also shouldn’t judge a research article by its abstract! If you like what you see in the abstract, keep reading! And proceed to the…

Introduction – (the “why”)

The purpose of the introduction is to provide context for the research study by outlining relevant background information. This section puts the study into context and establishes its significance and relevancy to the larger field. Not all scientific questions are worth pursuing; in fact, most aren’t. The introduction should answer the “so what?” of the study and help readers to understand why the research was conducted in the first place. A good introduction will typically end with the study’s hypothesis, a prediction, based on prior knowledge, about the relationship between two or more variables. The hypothesis is the crux of the scientific method and guides all research and experimentation. By the end of the introduction and before you continue reading the rest of the research article, you should be able to answer the following questions:

·  Why was this study conducted?

·  What questions are the researchers trying to answer?

·  What was the goal of this study?

This might take a bit of time to dissect, and I would recommend taking a highlighter to the article and actually underlining the answers to these questions. It will also be helpful to refer back to if you get lost at all while reading through the rest of the article.

Methods – (the “how”)

This is without a doubt the least exciting and densest section of a scientific paper. It is a description of the exact experimental procedures carried out in the study and it is meant to be exceptionally detailed. Why? Because one of the major tenets of good science is reproducibility. Reproducibility of data is essential for strengthening the credibility of the original research to ensure that the findings aren’t flukes or artifacts. The purpose of the methods section is to provide instructions for anyone to be able to carry out the exact same experiments from the study and, theoretically, generate the same results. What this means practically for the average reader, though, is that you do NOT need to read every word or understand every technique that is mentioned. Skim the subheadings in the methods sections so you can understand IN GENERAL how the researchers are testing their hypothesis to provide context for the results. Don’t get bogged down in the nitty-gritty of the exact procedures, unless you are planning to carry out a similar experiment yourself– which,  let’s be honest, most of us are not!

Results – (the “what”)

This is the meat and potatoes of any research paper; without a doubt the MOST important section. This is what the researchers found from the experiments they carried out to test their hypothesis. The main purpose of the results section is to provide a detailed account of the data that was collected, and the outcomes of the analyses performed, without interpretation or discussion of the implication. This is about as close to “unbiased” as science can get. At the same time, the researchers present the data in the order that they want you to read it, so try not to skip around. A clear and well-organized results section should read like a story and make it easy for readers to follow the progression of the study. Keep an eye out for words like “significant” and “non-significant” and try to connect everything back to the goal of the study and hypothesis.

You also might be happy to learn that this section of the paper includes pictures! Well, not pictures, “figures.” Figures mostly consist of graphs (bar graphs, line graphs, scatter plots, and/or histograms) that illustrate the relationships between variables, trends over time, and distributions of data. Other figures might include charts, photographs of raw data, microscopy images, flowcharts or conceptual diagrams, and maps or spatial representations of data in a specific context. Pay special attention to anything that changed relative to control groups (sometimes called “vehicle”, or “wild type”), as well as error bars, confidence intervals, and sample sizes. For each figure in the paper, you should be able to answer the following questions:

·  What was the rationale for doing this specific experiment/including this data?

·  What are the variables that are being compared?

·  What are the results of the specific experiment?

·  How does this information relate to the overarching goal of the study?

Discussion (the “where does it fit into the larger picture”)

While the results section is focused solely on presenting data, the discussion section of a paper is where the interpretation of those results occurs. This is where the researchers discuss and evaluate where their findings fit into the larger body of scientific knowledge that exists on a topic. In other words, this is where the meaning of the data is established. A good discussion section should connect back to the hypothesis of the study and examine how the experimental results align with or differ from what was expected based on previous studies and theories. Researchers will often acknowledge the limitations of their study in this section as well, and discuss potential sources of bias, methodological constraints, or other factors that could have affected the results. Remember, there is no such thing as a perfect study! Scientists revel in the imperfection of their work and are rightly exceedingly cautious not to over-conflate the implications of their work. At the same time, a good discussion section considers the broader implications of the study’s findings for practice and understanding with the field, and attempts to synthesize everything that was presented in the results sections into a coherent narrative.      

Extra tips for reading journal articles:

·  Make sure you have Google open when you’re reading an article. You will undoubtedly have to look things up as you go… and that’s okay! While you don’t have to understand EVERY single word of an article, there are certain foundational concepts you might need to review as you go.

·  There are a lot of acronyms in science. Pay attention the first time a new one is used so you know what it stands for the next time it comes up.

·  YouTube is an underutilized educational tool. If there is a concept or a method that you are interested in learning more about, type it into YouTube and find a video explaining it, or better yet, DEMONSTRATING it. But remember, not all YouTube videos are created equal! Rely on those deployed by people with the proper training and skillset to provide education on specific topics.

·  Similarly, ChatGPT is a good resource, as well, when used properly. Use prompts such as “Explain in situ hybridization simply” and it will walk you through the concept. If it is still too complex, prompt the AI to “simplify explanation” until the explanation is at a level you can grasp.

Final Thoughts

Reading scientific journal articles is a valuable skill that allows you to engage more directly with primary research. While it may seem daunting at first, with practice, anyone can learn to extract key information from these dense texts. Remember, the goal is to understand the main ideas - the why, how, what, and where of the study - rather than getting bogged down in every technical detail.

As you develop this skill, you'll become more adept at understanding scientific literature. However, it's crucial to note that being able to read studies doesn't equate to "doing your own research" in the fullest sense. Scientific consensus and expert interpretation remain vital. This skill should complement, not replace, reliance on those with advanced technical understanding of scientific matters.

The ability to read studies can enhance your scientific literacy and critical thinking skills, allowing you to better understand and appreciate expert interpretations and consensus views. It can help you engage more meaningfully with science communication and make you a more informed consumer of scientific information.

Remember, it's okay to find this challenging at first. With time and persistence, you'll become more comfortable navigating scientific papers, gaining a deeper appreciation for the complexity and rigor of scientific research.

Happy reading, and remember - every scientific paper you tackle brings you closer to being a more informed and scientifically literate individual!

Stay Curious,

Unbiased Science