How To Read Papers Effectively 2024 | Scanning like a Scientist

so scientists have to read over a 100 papers a day and to keep up with this literature is almost impossible for my last narrative review I also had to read up to 200 papers I believe and to keep up with this and to truly organize remember and reflect on everything you have read is the hardest skill to learn in this video I want to show you my technique for doing this and I have divided it into two sections first is to read as much about Fields you do not know about and the second part is

to keep up with the literature in a field that you are quite familiar with I truly wished I had this video at the beginning of writing my narrative review but I hope by giving it to you that it will help you a little bit with writing any type of review you're writing or with reading and keeping up with the literature in your field so let's get straight into it so the first part I want to talk about is reading papers in fields you do not know about and this is because I think that a lot

of us want to learn about new topics and we try to read academic papers to do so but find it quite hard and I think I've kind of developed a technique that can help with this and to allow you to understand a new field as much as possible through Reading papers so the first part of this process is paper selection so you need to ensure that you have papers that are actually good and that's because different from reading that you do at University is that papers are not always set in stone and they're just the

latest type of research but they're not necessarily true research or always good research so there are a lot of papers that are just not so good or even papers that were written 5 years ago that were trusted by the scientific community and that turned out to have a hypothesis that was disproven later so research is more of an iterative updating process than true knowledge set in stone and because of this fact you need to be quite careful about the research that you choose to read Because in research there are just a lot of facts that

were disproven so the way I do this is I usually ask my professor for the top 10 papers that he recommends in a new field and if you don't have a professor you can also look at this online for example through corsera there are usually a few papers recommended or if you're interested in Neuroscience on the neurom match website there are also a lot of papers recommended for specific Fields another thing you can do if you don't have access through this type of information is to look at a good review that was written I would

say in the last 5 years but maybe even in the last one or two years if that is possible and it exists out there so the second bit is the reading strategy and the reading strategy that I use is quite similar to reading strategy here that's called three bosses except that it's a little bit less extensive I believe but also have a look there if you're interested in common reading strategy that a lot of PhD students use so the way I go about reading a paper is I usually first look at the title the abstract

and the figures so I read the title perfectly read the abstract and then look at the figures and that's just to see is this paper actually in the topic that I want to learn about cuz sometimes a paper has a certain topic or a certain title that you think maybe suits what you want to learn about but when you read the abstract it's actually about something different so this can already help you kind of curate the reading list that you want to read afterwards I do read it sequentially in a field that I do not

know so for the second part of this video I will talk about fields that I do know about but for fields that I don't know about I usually do read the introduction M's conclusion and focus on the figures as well and this is because usually there are some terms that I am unfamiliar with and that I will need to look up as I'm reading through the paper so as I'm reading through the paper I usually highlight certain terms that I don't know so much about and I would look them up afterwards on and also something

that I personally really like to do is to kind of visualize the paper as well so I usually have the figures in mind and as I'm reading through it I keep referring back to the figures to see if I can understand what's actually being said so after you have read the paper in full it's time to kind of like dive into the math and the finer details and I usually use multiple resources for this because this is a field that you're probably unfamiliar with you want to use different type of resources than just the paper

because personally for me I find really hard to only read in a topic that I know nothing about and truly understand what's written so usually I listen to the top talks and you find these by just looking at the first author or the last author and kind of Google their name and then quite often there are talks of them in conferences for example that you can kind of listen to and understand a little bit more about their topic cuz usually that's introduced in a gor way during their presentation another thing I do is I find

related codes so for example a website is papers with code but also a lot of papers nowadays have at the bottom of their paper a section that is called code availability and there they list where you can find the code of their paper and I find personally if I have the time and I reimplement some of the bits that they've done that I understand it a little bit better also I would try to find relative uh relevant blog posts for example quite often for a little bit older papers you can find good blog post that

explain it in a more easy Manner and lastly I would also try to discuss it with peers so this is something that I've noticed that I find really nice as a PhD student so as a PhD student you usually have access to people in your lab that are interested in the same topic and I noticed when I really struggled with a paper I would go to one of my colleagues or a fellow PhD student and we would kind of sit down and maybe even explain little bits of the paper to each other and of course

if you cannot find this in real life it's also possible to find this on the internet to look for friends that maybe are interested in the same topic yeah and I think this is really where you go to the part of deeper understanding of the paper so through this deeper understanding you really want to focus on explaining the results to yourself without referring back to the paper so for example something that I do to test if I have a deep understanding of the paper is I would only have all the figures and try to explain

it to myself or to a friend or a fellow PhD student what I'm seeing in the figure and if I can understand the results that are being presented to me another thing that is really good to do during this point is to be quite critical of what you're reading so again research is an iterative process so no research paper is perfect it's always just a little test of a hypothesis that the authors have that they're presenting to you and you as a critical reader can decide if you agree or don't agree with a certain argument

that is being presented and I think a really good exercise for this part is to think by yourself what would be the next research question that I think needs to be answered for me to make this paper more convincing so this could be for example that you think that the sample size needs to be bigger you think that a different type of statistical test should be used or maybe you feel that the way they interpret their results are not fully sound with what you're actually seeing on the paper also something that I would consider during

this reading process is to read multiple papers at the same time so I personally refrain from just reading only one paper fully I I usually have about like five papers I would say on the same topic and I kind of compare how different authors write about the same topic and what kind of experiments they have done about the same topic and something that is quite interesting is that sometimes authors find the exact same results but they interpret it in different ways and that just shows you that as scientists we also have biases or hypothesis that

we preemptively believe and want to prove and sometimes these go a little bit above the results that we're actually showing and that it's not to say that these scientists are doing anything wrong but that they are trying to prove their hypothesis to you and you can choose if you believe this hypothesis or not so something that's very critical for me during this process is to have a good reference manager and that's where paper pal come in and they were so kindly to sponsor this video so I personally have always used reference managers and there is

a list of them out there but I think paper bell is very nice they have a really intuitive interf face and I think among all the reference managers it's probably the fastest to learn because it's already integrated fully with the Google environments so I'll show you really quickly how to use it so one of the things that you can do with paper pal is that you can search within paperpile itself which I find really nice so within the paperpile interface you can use their search bar and you have the ability to search online for books

and articles without leaving paper pal and you can also search not only for words or just the names of authors but you can also search for full phrases which I think is really nice if you're looking for certain phrases within a certain topic to add to your review for example the second thing you can do to find literature is to add papers through your library through the paperpile extension so the paperpile extension I will list down below you can download and then you can click on this PE button in your browser to save the reference

you can add the reference to folders and labels and add notes as you save the reference so for example I've done this for my latest review which had a lot of normative modeling papers and I just clicked and added it to my normative meling Library you can also save papers directly from academic databases which I find really nice so for example if you go to pet research RIT concess or Google Scholar which are all research bases that I use you can directly add it to paper pal like that and lastly if you work in Google

Docs you can for example write a small snippet of your research paper and then automatically already add your references as you're writing and I find this really nice because then you don't have to interrupt your workflow to think about how you're going to site the work that you're using so be mindful of sighting correctly and definitely tools like paper pal or other reference managers can help a lot with that so if you want to use paper pal to save you some time and keep all your referenes well organized I will list a link down below

so you can check it out so the second part I want to talk about is Reading in fields that you know about quite well and the reason for this is to keep up with the literature at least in my field I think there are papers coming out about like 10 15 papers a day and I set these kind of like research alarms in Google Al two keep me updated when an author that I like or a certain search term is coming up in a new paper and I think if you see the amount of emails

that I get with like an alarm that a paper has been published it's quite a lot let's just keep it at that and to keep up with this type of L you really need to be able to read papers a lot faster than you did in step one so I do think the reading approach for this then needs to be updated so the way I do it is with this socalled fast reading approach that I've kind of developed based on the three pass method but also based on my own knowledge and that is I start

with the abstract of course as always and I just read it quickly and see if the paper is interesting and here I'm quite strict so if there's anything that to me seems like a red flag I automatically remove it so for example is the sample size too small do they use a statistical test that I find finicky is there any type of terminology that I don't agree with and this maybe seems quite strict but these are true red flags for me for papers that in my regard then are not kind of like good to read

fully so the next thing I do is I look at the figures and I try to understand what the authors have done by just looking at the figures so usually I actually don't really need to read the introduction and discussion because because I already kind of know what the authors want to present and just looking at the figures gives me a more clear view of what their results are without muddling this with the preconceived bias that these authors have and the second thing then I like to do is I want to compare their results with

their methods so do their methods match with their results so sometimes what I see is that for example in their results they present something that looks intuitively quite nice so in their figures but when I actually go to the methods doesn't really match up or it doesn't really line so afterwards if I've seen the figures I've read maybe the results and uh method section then I sometimes do like to go to the discussion quickly to just see if their points of discussion match with the results that I've seen and to be very honest sometimes this

doesn't match at all so I've seen papers where the results were very flimsy they didn't really show True significance and in the discussion there are words being used like trending toward words significant very big red flag by the way um and this is just in general a little bit a gripe I have with the field and I think every field has its own pitfalls and I think that is something that's a pitfall in neuroscience and sometimes also psychology that um discussion or conclusion doesn't match the results actually being presented so be uh wary of this

so after I've done this something that I've done for my narrative review so a narrative review is not a structured review if you want to do a structured review find good YouTube videos on this or find a good article on how to do this because this is a lot more organized but in narrative review you kind of present the literature as you have found it and it's a lot less organized with the Search terms that you use but the way I've done my narrative review I want to show you quickly is that I used an

Excel sheet so this is the Excel sheet for all of the research papers that I read and I in this sheet noted down elements of the paper that I might want to reference later and this really helped me to see overarching teams and remember the differences and similarities between papers I and I actually learned a lot doing this and I'm considering doing this for every paper I'm going to write in the future because quite often you read over 50 papers and it's really hard to keep a clear note of which results belong to which paper

and also which paper you can or should site when you're looking at different results also something that I like to add as another column is this question of the follow-up paper that I think these authors need to write and usually when I can answer this question it does show that I quite understand what these authors have done and also what I think is missing so it allows me both to kind of get an overview and also to indicate that I've read the article critically so the last thing that I wanted to end on is some

tips for Effective reading that I've gathered over the last few years first of all some tips for making reading a habit so so I really find it truly important that you have reading as a habit or something that you do daily something to do this is to make a dedicated reading time so possibly a specific time during the day that you will always read one or two papers and maybe even put it in a specific location like a library also make it enjoyable so make your favorite drink for example a little coffee as you're reading

and also something that you can consider is to have a reading group so as I said you can gather together a couple of your student students or a couple of students to make it into a reading group for discussion and accountability the second thing for Effective reading is to also read for creative ideas so I do think reading papers is very important but I also think reading outside of papers is very important and this for me actually taught me to speedread a lot more and to also enjoy reading a lot more so something that you

can do is to read outside of Science in books that are maybe adjacent to science so for example you can consider pop science books I will list a few here that I personally really liked reading and through reading these pop science books although they're not always accurate I do think they teach you a way of speaking about science that's not so dry and not so formalized also something that you can consider is to write these little mini essays about topics I'm really considering about making a video about this as well but I think if you

write a little mini essay about something that you're learning or something that you're interesting interested about you will remember it a lot better and that's something that I've been enjoying doing for the last few weeks and that I also want to start posting on my blog that I will list down below so these were some tips for reading in 2024 that I've gathered and kind of Quantified over the years I'm always updating the way I read and I always try to learn new skills and I still don't think my reading techniques are perfect so if

you have any tips for me I would love to hear them so put them down below also if you're interested in not taking and reading I do think this video is maybe a good follow-up video and otherwise see you next week bye