How to read a scientific paper

[music] Welcome back in this Part 2 episode on Scientific Literature. Professor Aaron Frazier and I are going to look at how we would actually approach reading a scientific paper. It's not as simple as going from the beginning to the end. So I guess the important question that is, if I was to hand you a scientific article or give you what a mind to read because I know you read everything I write there. If I handed you a paper, the big question that I get from students and it's an interesting one is do you start at the beginning and read to the end or do you use something different? Yeah, that's a great question and I get that question from students as well because it certainly is kind of what we're trained to do as readers when we're presented with something and is to start at the beginning and go to the end. And certainly there are lots of documents for which that is absolutely the main way that you would want to read it. Like I'm picturing by a mystery novel and choosing not to read it from the beginning to the end. We're very much trained to do that but actually I think I would hazard a guess that most people who regularly read articles or manuscripts that are within the scientific literature probably don't typically do that. Okay, so what would you read first? Yeah, so probably the first thing I would read, like I guess what I'm thinking is I find myself reading in a couple of different contexts, the scientific literature. So sometimes I'm just doing it to try to stay current and learn new things. So I'm seeing that a journal has published a new issue and there are a list of articles in it. So maybe I'll choose to read some and not others. So I might be making the choice to read for that reason or I might be really actively on the hunt, right? Like I might be planning a project or writing my own manuscript where I need to see what other things have happened so that I can reference that work and what I'm writing in which case I might be like using databases through the library to track down literature that is relevant to what I'm doing. In both cases, the first thing that I'm going to see is the title, right away. I'm going to see what is hopefully a relatively descriptive title, descriptive enough that I get an idea of what the work is about. And then I also am going to see the abstract. Oftentimes if you're actively on the hunt for literature, so like if you're using a database, often if you see a list of potential journal articles, you'll just see the titles. But if you click on the title, what you'll see is the title and the abstract. So that gives you a chance to read the title, read the abstract, and then decide like, yes, this is something that I want to go ahead with or no, I'm not going to read this right now. It's not what I'm looking for. So title and abstract right away. And like you said, it's kind of the preview for the paper. It gives you an idea of what you're going to, what you're going to find. Do you have a next step that you would go to? Here it's where I think what I do would really depend on why I was doing the reading and the extent to what what I was reading is within my field. So I think if I'm reading something that is in a field that I'm really familiar with, I'm probably not going to start with the first sentence of the introduction and just work my way linearly through it because if we think about what the introduction is for, it, you know, it provides a lot of background, provides a lot of context. And I might kind of be pretty familiar with that information already if I'm reading something within my own field. So this is not to say that I'll never read it. And in fact, I would like to point out that in fact, if I'm using a paper, I will definitely read the whole thing eventually. But the first thing that I'll do as I'm sort of just going through it to see what's going on is I might go to the end of the introduction and check out that statement of purpose research question objectives hypothesis predictions. That kind of thing. So that will really help me to see what the study was all about. And then I find that I often will go and look some of the the figures that are associated with the results. And like you were saying earlier, often figures with their captions are designed so that they can stand alone. Even if you haven't read the whole paper, you are able to look at the figure, read the caption and already get a snapshot of probably some of the most important take home messages from the study. So I might have a look at some of the figures in the results. And then I often find it's really very handy. Like we talked about how many discussion sections do start with a statement that highlights the main points or the main important points from the paper. And I'll often read the beginning part of the discussion because that's often where the authors will explain what they think are the most important take home messages from their work. And the same with the the conclusions at the end are also often a good place to start. I wouldn't want to stop with reading those things if I was wanting to really use the manual or the paper in a meaningful way. Like if I wanted to cite it or or plan my own work building on on that work. But in terms of just like trying to figure out what is this paper about, those are the places that I would look first. I think I'm remarkably similar. So abstract and then probably figures to see what they found. And then I might go back to that last paragraph of the introduction. So I think I do the same thing. I think it's really counterintuitive to say, well, I just randomly pick these paragraphs and I read them out of order when somebody has sat down and planned a paper in a particular order. But I totally agree with you that it you don't have to read them for beginning to end. You can jump around in a paper with the bit you need. And interestingly, neither one of us said methods because I almost never read the methods until the very end. What is the story you're telling me? What is it that I want to take home from this paper? Later on, I'll go figure out how you did it. But that's probably less interesting than to me than whatever conclusion you've drawn. And the other thing that honestly, it's funny. We're saying that, oh, you find people are not necessarily all that interested in the literature cited or the references section. Because I often find that I can spend quite a bit of time, especially if I'm just getting familiarizing myself by doing reading within a field I maybe don't know as well. I may spend a lot of time looking at the literature cited section of some papers because they can help me to see the scope of work that already exists. And I can see if I've missed any really important contributions as I'm trying to do my own little survey of whatever little area of a field that I'm trying to learn about. So I find the literature cited section could be really, really useful in that respect. I agree with you, and I would definitely go there looking for more information on a topic. If the article I'm reading has felt it was important enough to link it to some other article, then that one might be key to me as well. Although you can't really read it. It's not like text you can read. It's more like a treasure hunt trying to find a bit inside that's important. I feel like all of the things we've talked about so far, that's how I would approach something like a practitioner within my field where I'm trying to figure out what is something about, how does it relate to other work I've read, maybe other things that I've done, does it seem like it's going to fit with the thing that I'm writing right now. I would say that if I'm coming at a journal article in sort of a more of a naive headspace, like I really don't know much about it, I really don't know much about the field. If I am just reading to be interested and to learn more, I might be more inclined to read it the way that it, just the way that it's been presented. Because I probably will find that introductory information really useful because I maybe don't necessarily know the background, the terms, the knowledge gaps, the context, depending on the perspective from which you're coming to do the reading, as well as why you're doing the reading and what your background knowledge is for me anyways, really impact how I go at something. I think the last thing I want to ask is this moment of total honesty here, which is when you're reading an article, I don't know about you, but I often run into words I don't know. For sure. Topics I've never heard of, I think there may be this assumption that we don't struggle to read these things. Is that true of you too that you're always stuck up? I mean, I Google words all the time to find out what is that word mean? I've never heard that before. Yes, absolutely. A hundred percent. I'm always having to look at things up and often I'll find myself googling methods as well or statistical techniques if they're not ones that I've used. Yeah, often you need to do a fair amount of like work to really understand what you've read. I think what is behind that is that there's a lot of emphasis in scientific writing on being very succinct. So unlike some other forms of writing where it might be a little bit more wordy, there might potentially be more descriptions of terms or that there just may not be space for that in these articles. know from the perspective of writing these kinds of articles, there's often, you know, a lot of thought goes into how you write them and what you say to make them as broadly understandable as possible to your audience. But nonetheless, they're, I'm sure, are still things that need to be looked up. Oh yeah, I learned a word last night reading that article, which is, but I think that's important, is that we don't just read them and understand them. We might have to read them two or three times for different levels of detail. One last question is there's not to go into detail on this, but there's a totally different kind of paper that you'll find in the same journal and it's the confusing bit. So what we've been talking about is what we call primary literature. It's the actual experiment, the method and the result written by the people who did the methods and the results. There's another kind of paper that appears in the same journals, which is not really primary literature. It's called a review. What is a review paper and what might it be useful for? So review papers usually are trying their objective, their reason for being, is that the authors are bringing together and reading and summarizing and synthesizing information from ideally all the literature within the scope of whatever it is that they've written their review about. Sometimes reviews are the way they're done is extremely rigorous, like sometimes they actually will have a little method section where the authors will say, you know, the way we found papers to include in this review is we went on these databases and searched these terms and any paper that came up, we read it and have extracted information from it or have summarized it. Sometimes they're a little less rigorous than that. They might not necessarily have those specific methods, but the idea or the sentiment of them is that still they are bringing together a big quantity of literature within a section of a field and usually the paper in its entirety does provide a summary. So they're often is just a lot of information that is just summarizing the main take-home messages from all of these different studies, but there usually is a synthesis component to a review as well. So the idea there is that in reviewing and reading all of this literature, the authors of the review may find that certain information is starting to emerge from reading many, many studies. So they might start to see that many of these studies are having similar findings or in fact, maybe not at all. Like maybe it's all over the place people are finding all sorts of different things or maybe there are big knowledge gaps, like maybe there's a lot of research in one part of the field and not very much in another. And so these are the kinds of things that you can say if you're looking at all sorts of different papers within that field. And sometimes these assertions are backed by data. Like sometimes when people do a review they'll actually try to collect a specific types of data from every paper that they read and then they can they can summarize those in a table or even in a figure. It does sort of provide like just a state of the field at this moment in time. So it can serve as a way as a vehicle to inspire people to do research in different directions. Like oh this is also interesting that people have found this, but we really don't know about this or here's a here's a knowledge gap that could be addressed. So that can be really useful for the authors and also for the people who read the paper. I also find for myself if I'm just starting to learn about an area for the first time a review paper can be an excellent thing to read because the whole point of it was to bring together most or all of the relevant literature about a specific topic and summarize it. So it's kind of like your one stop shop for lack of a better word to get you started on learning about a new topic. Also if that's where a literature cited section can be really really useful. If you go if it's a recent review paper and you look at the literature cited section you can find a pretty comprehensive list of studies that are relevant to whatever the topic of interest is. Up the top of my head those are the main functions of a review. Can you think of others? No I was going to say the same thing it's usually where I start. If it's a topic that I want to learn about that's really new to me I'll look for a review paper first because they're probably going to summarize huge portions of the literature and save me time reading to do that work for me. Have you ever written one? Yeah yeah I have. Yeah me too they're hard work. They are hard work actually. Yeah I mean anytime I do it actually by the time I'm done I think wow. Usually it ends up with me having made it an Excel spreadsheet with rows and rows where every row is summarizing a different paper. Okay thank you very much Aaron. I always ask this question when I'm talking to people doing this. How did you become a professor and was that what you intended? Well that's a great question. So how did I become a professor? Well I actually for those of you who are listening to this so Dr. Claire and I go way back we've known each other for close to 20 years now and we actually met because we were both students doing research in Dr. Brock Fenton's lab at the University of Western Ontario and London Ontario and I did both my master's degree and my PhD at Western University. I also worked with Dr. Fred Longstaff. I finished my PhD at Western and the whole time that I was doing the research component of my graduate work I also was really working hard to develop my university teaching skills. I really love teaching it's something I've enjoyed my whole life and I was very lucky at Western because I was able to work in their teaching support center there where I did a lot of training on techniques and strategies for teaching in higher education. I was involved in the teaching assistant training program there and so my goal at the end of my PhD was I really was hoping to get a position as an assistant professor and I was particularly hoping to work at an institution that had positions where there was a fair bit of undergraduate teaching involved that something I've always been interested in and that I was really looking for so I was really excited to see a job posting at Grenfell campus which is where I currently work for a position they were looking for a vertebrate biologist who could have a research program but who also would do quite a bit of teaching each year and I applied for that position and in retrospect know I was really fortunate because I was able to get that position fairly shortly after I finished my PhD and I've been there ever since. Thanks again to Professor Aaron Frazier for joining us on the BioAudio podcast for this two-part episodes on how to read scientific literature. This has been a presentation of the BioAudio podcast. I started BioAudio as a live Q&A session with the class when they had questions that were outside my area of expertise. Over the course of a few years live sessions became some recorded sessions and then a hosted interview and then some audio files for the class. At the request of some of my students I made them public as a podcast so they could more easily listen on their phones. I was not prepared for how enthusiastic the class was and a few episodes soon became a dozen and then enough to provide a free alternative to traditional textbook readings. The goal is to learn through interviewing experts and former students and to make an alternative free and more inclusive resource. We are not perfect but we're learning as we go. If you have enjoyed this episode particularly if you are a student leave a review on Apple Spotify or wherever you get your podcasts. Follow me on x@doctor_bat_girl on Macedon at [email protected] or blue sky at profbatgirl.bsky.social where I post new episodes and new news from my research lab. I hope you've enjoyed this presentation of the BioAudio podcast.