Table of contents
1. How do you go about reading a paper?
2. What should you do if you encounter something you don’t understand?
3. How to overcome frustration when reading a paper?
Even professional scientific researchers cannot read various research papers like novels - many researchers say they encounter a lot of confusion and frustration when reading papers. Therefore, Science invited a dozen scientists from different fields and with different qualifications to tell us how they do it.
Although reading scientific papers becomes easier with experience, the difficulties encountered along the way are real, and each researcher needs to discern and choose the method that works best for them.
1. How do you go about reading a paper?
Cecilia Tubiana, Scientist, Max Planck Institute for Solar Energy Systems, Göttingen, Germany
I start by reading the abstract and conclusion to get a general idea. The conclusion section helps me understand whether the goals outlined in the abstract have been achieved and whether the work described will be useful for my own research. I also pay attention to figures and tables because they give me the first impression of a paper. I then usually read the entire article from beginning to end, going through the sections in order so that I can follow the workflow the author is trying to convey.
If you want to be effective at reading papers, you first need to have a clear idea of what kind of information you need to obtain, and then focus on that aspect. You can compare your conclusions with results provided by the authors, put your own analysis into context, or expand on it with newly published data. The reference list tells you what colleagues who have conducted similar research have done, and gives you a first impression of how to use this paper. This can help you determine whether this paper is highly relevant to your own research.
Jeremy C. Borniger, Ph.D. Candidate, Neuroscience, The Ohio State University, Columbus
If it's just to get the main points, I'll read the summary, then jump to the diagrams, and skim the discussion section for key points. I think figures are the most important part of a paper, because the abstract and body of the paper can be written into a compelling story (but the figures represent completely objective research results). Then, if there's anything unclear to me, I go to the methods section.
If I want to delve deeper into a paper, I usually read the entire article and then read a few articles written by others on the same topic.
Then, if the authors' research is similar to my own, I'll see if their data are consistent with our findings. If there is an inconsistency, I'll think about what might be causing this. Additionally, I would think about what our own models would look like if we used the same approach and what we might learn from it. Sometimes it is also important to note the author's motivation for using a particular experimental method. Did the authors use a fuzz test instead of regular analysis, and why did they do this?
Kevin Boehnke, PhD Candidate, Environmental Health Sciences, University of Michigan, Ann Arbor
I usually start with the title and abstract. This lets me know whether I am interested in the article and whether I can actually understand it - both scientifically and linguistically. Then, I read the introduction so that I can understand the problem being described in the article. Next, jump to the charts and graphs to understand the data. Finally, I read the Discussion section to understand how the paper fits into the general body of knowledge.
I will pay attention to the authors' statements about the limitations of the study and the reasonableness of the derivation of the data. Some people take their claims very far, which might be a "red flag" to me. I will also step into the role of epidemiologist to ensure that the experimental design of the article is truly suitable to test the hypothesis that the authors want to test.
Gary McDowell, postdoctoral fellow in developmental biology at Tufts University in Medford, Massachusetts, and visiting scholar at Boston College
My reading strategy depends on the paper. If it applies directly to my current topic, I will read the entire text carefully, except for the introduction, which I may already be familiar with. But I always want to find if there is anything in the article or figure that deserves my special attention, and then I will read the relevant information in the Results and Discussion sections.
I also check the references to see if there are any articles that might be of interest to me. Sometimes I am curious to know who in the field is cited by others (or who is not cited by others), and whether the authors choose to ignore certain directions in the field. I often find that supplemental figures actually provide the most interesting parts of the results, especially when the results relate to parts of the paper that the authors did not cite, or when the results are unclear and do not support the overall story the authors are laying out.
Lina A. Colucci, PhD Candidate, Harvard-MIT Health Sciences and Technology
When reading a paper, setting a task for myself to read and write at the same time can help me become an active reader, instead of letting my eyes stare blankly at a mountain of text, reading one paragraph and forgetting another. For example, when I read background information on articles, I save the information on a specific topic in each article in a Word document. I'll write some comments about new ideas I've gained, or questions that I need to explore further. Then, in the future, I'll just have to read this document instead of re-reading those papers.
Likewise, if I'm trying to figure out how to do a particular experiment, I create a handy table in Excel that summarizes how different research teams conduct that experiment.
Lachlan Gray, Associate Director, HIV Neuropathy Laboratory, Burnet Research Institute, Adjunct Research Fellow, Department of Infectious Diseases, Monash University, Melbourne, Australia
The Results and Methods section of your paper allows you to “bug” your article to ensure it meets scientific rigor. You need to pay attention to the types of experiments performed in the article and consider whether they are most appropriate to address the question the author is asking. Make sure that the control group and experimental group set by the author match and have sufficient sample size. Often, conclusions are made based on a limited number of samples, which limits their significance and importance.
I like to print out papers and highlight the most important information, so after a quick re-read, I can recall the key points. The most important ideas will change your thinking about your research topic or give you new ideas and directions.
Brian Nosek, professor in the Department of Psychology at the University of Virginia and executive director of the Center for Open Science in Charlottesville
I choose to read papers that are relevant to my research area and will generate interest and discussion because they are pushing us to do psychology or scientific research in new directions. Mostly, I focus on the paper's methodology, experimental design, and statistical analysis. To me, the most important part is first what the authors did (methods) and secondly what they found (results).
It is also interesting to understand why the authors did this study (Introduction) and what they think of the results of the experiment (Discussion). When I get to know the field a bit, I usually don't care much about these parts. However, when I know little about this field, I will read these carefully, because then I will learn a lot about the assumptions of this research field and the methods to test them.
Ulf Leonhardt, Professor of Physics, Weizmann Institute of Science, Rehovot, Israel
First, I'll read through it quickly - just to see if I'm interested in the paper. If I'm interested, I'll read it again slowly and pay more attention to the details.
If the theory of the paper is crucial to my research, I will re-read the paper. In this case, I only use the beginning of the theory in the article as a starting point, and then study the rest myself instead of reading the article. Sometimes this is a very painful and slow process. Sometimes I get annoyed when the author is not clear enough, misses important points, and leaves unnecessary nonsense. Sometimes I get inspiration and energy from papers.
Marcia K. McNutt, editor-in-chief of science magazine
As editor-in-chief of Science, I must always read and understand papers outside my own field. Generally, I would start with a summary from the editor responsible for the paper. I am interested in all scientific disciplines but only have a deep understanding of one area of research, so this type of summary makes sense to me. Next, I’ll search to see if anyone has written news coverage of the paper. Then, I want to see if other scientists have expressed their opinions on this paper. The main goal of "opinion" is to extend the information conveyed in the paper, but the author of "opinion" is also able to extract the essence of the article very accurately, which provides great help to non-professionals.
Next I work on the abstract, which is usually written in a form that is broadly acceptable to the journal's readers. Finally, I start reading the paper itself. I read the introduction, conclusion, look at the figures, and then continue reading the paper.
2. What should you do if you encounter something you don’t understand?
Jesse Shanahan, M.A. Astronomy, Wesleyan University, Middletown, Conn.
If it's just a part of the article, I'll record it and look at it later. If I really find it difficult to read the article, I will try to consult a review or textbook to get the necessary background knowledge, and I often find that I am more efficient this way.
There are a lot of abbreviations and jargon that may belong to specific subfields, so I usually don't go into too much detail unless it's for my own research. But I always try to take the time to figure out the method used in the article.
Jeremy C. Borniger, Ph.D. Candidate, Neuroscience, The Ohio State University, Columbus
I usually stop immediately and look up the parts I don't understand. If I don't understand key words or terms, there may be no point in reading further. However, this can be counterproductive, as I tend to get stuck in a cycle of (what is X? Oh, X affects Y, so what is Y? etc...). It can be fun when you understand how everything is related, but if you're interrupted, you won't be able to stay focused on the current task.
Sometimes, all the terms in a paper may at first glance cover all the key points of an experiment. In this case, we must ask ourselves: "What question did the author want to answer?" Then you can determine whether they succeeded or failed.
Cecilia Tubiana, Scientist, Max Planck Institute for Solar Energy Systems, Göttingen, Germany
It depends on how much the incomprehensible characters prevent me from understanding the main idea. When I first read a paper, I usually don't try to understand all the details. If the incomprehensible part is important to my research, I will try to ask colleagues or even contact the lead author directly. The last resort is to go back and consult the paper's references to get all the background information, because time is limited and collaboration and personal contact can more effectively solve specific problems.
Gary McDowell, postdoctoral fellow in developmental biology at Tufts University in Medford, Massachusetts, and visiting scholar at Boston College
Sometimes you can read through a paper and any terms you're unfamiliar with will become clearer by the end. If there are a lot of terms you don't understand, stopping and looking for explanations elsewhere is often the way to go. I will use Google to quickly search for the article's ideas, themes, methods, terms, etc. If these terms are very dense in the article, it will take some reading to begin to understand them.
Brian Nosek, professor in the Department of Psychology at the University of Virginia and executive director of the Center for Open Science in Charlottesville
I ask myself a question: "Do I have to understand what they mean to get what I want from the article?" Now, the articles I read in my research area are well beyond my expertise, and I often don't need to A superficial knowledge of its content. If I couldn't read the paper without understanding the underlying content, I would do more background research.
3. How to overcome frustration when reading a paper?
Brian Nosek, professor in the Department of Psychology at the University of Virginia and executive director of the Center for Open Science in Charlottesville
all the time. If the paper is about a problem I'm grappling with, there must be something crucial in the paper that I don't understand. This confusion is not a threat: it is an opportunity. I'm ignorant - I need to become less ignorant - this paper might help me.
At the same time, some articles are poorly written and not worth your time. Others have definitely written the concepts more clearly so that I can focus on understanding the gist rather than the syntax.
Jesse Shanahan, M.A. Astronomy, Wesleyan University, Middletown, Conn.
I feel particularly overwhelmed if the article goes beyond my research area, is long, and is filled with jargon. In this case, if possible, I break it into chunks and read it over a few days. For really difficult papers, sit down and discuss them with colleagues.
Kevin Boehnke, PhD Candidate, Environmental Health Sciences, University of Michigan, Ann Arbor
Yes, many times. That’s why I developed my own reading strategies through talking to other scientists and through trial and error. Sometimes I throw up my hands in frustration, throw away those troublesome papers, and never read them again.
Jeremy C. Borniger, Ph.D. Candidate, Neuroscience, The Ohio State University, Columbus
Yes, in this case you have to realize that some papers are the result of dozens of scientists working for many years. Expecting to digest and understand everything in a paper in an afternoon is unrealistic.
Cecilia Tubiana, Scientist, Max Planck Institute for Solar Energy Systems, Göttingen, Germany
I often feel overwhelmed! But some parts may not require as much depth as others. You need to know your limitations: are there parts of the paper that you would like to emulate but are not within your expertise? Can these parts be made “understandable” through collaboration with others?
Gary McDowell, postdoctoral fellow in developmental biology at Tufts University in Medford, Massachusetts, and visiting scholar at Boston College
If I feel that the paper is very important to what I am doing, I will put it down and come back to it several times later. But if it's too overwhelming, then I have to put it aside unless one of the colleagues I'm in contact with can explain it.
4. Any other suggestions?
Lina A. Colucci, PhD Candidate, Harvard-MIT Health Sciences and Technology
If there is a groundbreaking paper that I want to fully understand, I will try to find an opportunity to give a report to everyone at the group meeting. Discussing a specific paper and answering questions in an oral presentation is the best way for me to learn this material.
Also, find a good reference management software. I use Mendeley software to help me do research, read literature and write papers.
Rima Wilkes, Professor, Department of Sociology, University of British Columbia, Vancouver
At first, beginning paper readers will find that their progress is slow because they do not have a frame of reference for what they are reading. But we can create a "brain library" through reading, and after a few years it will be easy to insert papers into your brain's bookshelf. This way you can quickly scan a paper to understand its contribution.
Kevin Boehnke, PhD Candidate, Environmental Health Sciences, University of Michigan, Ann Arbor
Be patient. Don't be afraid or embarrassed to use Wikipedia or other outside resources, such as blogs, to learn about your topic. Remember to ask questions. If you don't understand the paper clearly, talk to people in your circle. If you are still confused and it is important for you to understand these concepts, please email the author.
Marcia K. McNutt, Editor-in-Chief, Science
Don't hesitate to talk to a more experienced scientist. You are doing them a favor: they have the opportunity to explain to you what a complex paper means in terms you understand. All scientists need experience communicating complex concepts using common vocabulary.
Jesse Shanahan, M.A. Astronomy, Wesleyan University, Middletown, Conn.
If possible, read often. Try to summarize the article into a table of contents document that includes key information with references and may also include one or two figures. Pay attention to how different articles are constructed and articles with different writing styles. This will help you develop an effective and unique reading style.