Speaker 1: Performing a literature review is an important step in conducting research, as it allows you to gain a comprehensive understanding of the existing literature on a topic. So hi, I'm Charlotte Vreese, a third-year PhD student in computational neuroscience, and today I want to talk about my literature review workflow. So I have had to read a lot of papers over the last few years, and to keep a comprehensive book or a comprehensive system that allows you to synthesize all these papers is actually quite difficult, and I think I've finally kind of got there. So I want to introduce you to my system such that you may also benefit from it and hopefully read a lot more papers for your PhD or if you're a student for your research work. So let's get straight into it. So there are six steps that I want to introduce you to, and the first one is to define your research question clearly. I think if you dive into research without having a clear research question, it's actually really overwhelming the amount of research that you will find and the amount of papers that you will find. So usually it's quite good to already know beforehand what kind of research question you would like to dive into. So for example, today we're going to take the example of the Bayesian brain. So if you want to learn more about the Bayesian brain, your research question could be, for example, what is the Bayesian brain, what does it mean, how does it work, etc. So the second step is to find a good review and find the accompanying relevant research on Scopus. So I personally like Scopus the most, as I think it's the most comprehensive knowledge bank right now on the internet. But there are a few other knowledge banks that I will list down below, and on these type of websites you can find relevant books, articles, and reviews. But the thing I like most about Scopus is that you can also list the articles based on highest citation, for example. The authors that wrote it, when it was written, and I find it pretty intuitive how the paper system is structured. Whereas for example on Google Scholar, even though I sometimes do use it, it is sometimes a little bit a black box how they organized the papers. So in Scopus usually the first thing I do if I'm new to a topic, I find a review. So you can just google like what I'm doing right now, so Bayesian brain review, and then you search. And usually I list it by the highest citations, if I don't know that much about the field yet. But this does mean that newer articles go more to the bottom. And then for example this one, a Bayesian brain, the role of uncertainty in neural coding and computing. So I would probably select this one, and then try to find the paper. And another thing that I do like to do is that I do like to see in the field which authors have published the most. So this you can do on Scopus, you can see which authors have published the most. So for example for the Bayesian brain, Carl Fristen has published a lot, which I also already knew. But if you're new to computational neuroscience, look up Carl Fristen, he has done some amazing work. And I think by also looking at the authors, then you also know which people you should follow, which are the most contributing authors in this field. So then the next thing that I do after I have found the articles that I like, and that I want to read a little bit more, then I usually import them in Zotero and start highlighting and reading in Zotero. So the thing you want to do with highlighting and reading is that you want to look for patterns, you want to think of questions, you want to see what kind of open gaps or gaps in the literature there are. And I have made a video about how I usually go through certain scientific articles that you can find right here. So for example if I have this paper that I want to read, so for example in this case predictive coding, an account of the mirror neuron system, then I click on the add-on from Zotero and you can select here the map that you want to save it to. So usually I have for every paper that I'm writing, for every topic that I'm researching, and usually also nowadays for YouTube videos, I have a separate map on my Zotero that I link all the articles to. So I save it automatically in this way to this map. And what you can then see in Zotero is that all your papers are listed here. It usually also automatically downloads PDF, which I really like, if it's available. If it's not available you sometimes have to add it. And then the thing that I do is I go over the paper and highlight all the parts that I find interesting and that I want to reference. So this for me is really where I do my rough reading, where I do my first reading, and also where I collect all my papers. So I highly encourage you to make different folders for different projects that you have, because I think at the end of your PhD you will have probably read over 200 papers. And if you just all have them at the same place, like what I did at the beginning, it becomes really unstructured super fast. So then the fourth step that I do is I open Research Rabbit and I import my Zotero into Research Rabbit. So the thing that I like about Research Rabbit, you can see it here. So for example, these are papers that I read in 2022. This is a map I make. And for example, here's the history and future of the Bayesian brain. And then you can look for similar works. So you have here all the similar works. And the thing I like about it is that you have this kind of graph-like structure. So you can really see what important notes are. So for example, Karl Friston wrote a really important paper, Lily Crabbe, and also these others. And I think by looking at it this way, I usually uncover the key papers in certain research areas or certain topics, which sometimes is a bit hard to see if you are just listing the papers in Zotero. So for me, this is really a good way to get a visual overview of all the material that has been written in the field. So the next thing I do is I organize it in Notion. So all the highlights that I've made, all the things that I've thought about, it usually can be synthesized down into one question. For example, one statement, one main fact that I want to remember. And usually I put this in the project that I'm thinking about. So I've made a video already about how I use Notion as my second brain. And I really think having all the main or key points of every article in one place, for me at least, is really beneficial. And I do know sometimes this is a little bit of a hassle to transfer this. So I have thought about like an automatic transfer, but I'm not still super satisfied about it. So if you have a way that you automatically link your Zotero to your Notion, let me know in the comments down below. And the last step, step six, is to really synthesize the literature into your own review. And this is not something that I always do, but sometimes your professor asks you, for example, to write a review paper, or you have to write a certain part of a review for an article or a course you're following. And I think actually this part is the most interesting, because acquiring all the knowledge, although it takes some skill, it's not really an inherent ability. But actually thinking about the knowledge, really interpreting it, seeing where the gaps lie, thinking of your own questions, is where it becomes really interesting. So if you want to know how I do it, let me know down in the comments. But I do think in general, just thinking about it for a moment and sitting down and writing your thoughts down after you've read all the literature already will take you quite far. So of course doing a full literature review is a really time-consuming process, and I also realize that. But I do think it's a key part of your PhD that I also sometimes forget about, because when you're doing research and when you're thinking about thoughts and everything, you sometimes forget that there's already so much knowledge that has been written about a certain topic. So I hope that my tips helped you a little bit, and that my workflow will speed up your process of doing this quite tedious work sometimes. And let me know if you have any tips for me, I would love to hear. So put it down in the comments below, and otherwise see you next week. Bye.