How to Research Any Topic - Deep-Dive like a PhD Student

Live&Learn
1 follower
1 Video View·Feb 15, 2026
So scrolling is easy and doing deep research is not and that's why on average we spend such a
long time scrolling on our phone instead of working on that essay that is dusting off
in the corner. Doing deep long hours of research might seem the exception and not the norm,
however there's something really meditative about the research process and I think if we
learn a little bit more about the research process this can actually help us in daily life
to become a little bit more observant. So I've been doing research for almost 60 years now as
both a PhD student and now as a postdoctoral researcher and I kind of want to lift the veil
of how this research process goes. So how do we come up with a good research question? How do
we find the right literature resources? How do we know what we can trust or shouldn't trust?
And over time I think this process will become a little bit more intuitive because I've kind
of learned when I told people about the research process or how I would deep dive into specific
topics that it's actually not as intuitive as I thought. So I thought I might tell you a little
bit about how I do research or how I would tackle a specific question or problem that I face both in
daily life and in my work. So let's get straight into it. So in general the first thing of course
when you do research that you want to do is to pick a topic and here I would advise to start
from general to specific. So most of us have a topic in mind that we might want to learn something
about so that can for example be neuroscience, the brain, consciousness, mental health. But this is
very general and if you want to do research about a specific topic you actually need to be a little
bit more specific and it's really helpful to make a very specific question on the type of research
you want to do. So for example you first want to narrow it down to one specific angle. So if you
want to learn a little bit about mental health you might want to narrow it down to the impact
of socioeconomic status on mental health. Such that when you do literature research you are not
just googling mental health question mark and then you can narrow it down even further. So for example
what is the impact of low socioeconomic status during childhood on later development of psychiatric
disorders? And you already kind of get a feeling that this question will have a specific answer as
opposed to mental health as a general topic or just socioeconomic status and mental health. So the
more specific you make your question the better your literature review basically will go and the
better resources you will find over time. So write down a really specific research question and take
time for this and it can also even be that as you're doing your literature search, which we
will talk about a little bit later, that you will refine your question over time. So then the next
step you want to do is what I call the literature hunt and this is really a very involved process.
So it will take a lot of time. So as I explained it, don't get scared and don't think like oh my
god that is way too much work. But over time you will get quite good at it and I think even though
it sounds like a lot, I think you get quite fast at finding the right resources. Because it does,
yeah it is a specific skill to find good resources. So in general what I usually do when I
have a specific research question in mind is I or start with the classic papers in that research
field and that depends a little bit on the research field you are in. If there are classic
papers or I would start with a recent review of the last five to ten years and again that depends
a little bit on the research field you're in, how fast it changes. So in general when I find a good
review it's quite hard to decide a journal that has good reviews. So for example I know for
neuroscience that for example Nature is quite good or AGP is quite good and I've sometimes
gotten the question as to how to decide if a journal is good in general. And I don't really
think there is a way to decide this. The reason I trust specific journals is because specific authors
that I trust and that have published quite good work over time publish in those journals and it's
not necessarily that those journals are better than other journals. One thing to always look out
for is if a journal is peer-reviewed but most of the reputable journals are. And then I think you
want to see where are authors publishing that you like or that you think their work is good. And
this really comes with experience so it's quite hard to skip this step. But in general you can of
course rely on a bit the more yeah the bigger publishers such as Nature for example or AGP.
But over time you will also get a feeling which journals are trustworthy and which ones are not.
So then you have your review that you can start from. I would immediately go down to the citation
section or the reference section where they basically put all the citations of the work they
think is relevant for the field. And then you already have a goldmine of information. So all of
these papers I would put in a common reference manager. This can be Zotero, Paperpal or anything
you want. And then I would slowly try to kind of make maps between which authors are the most
important and which papers are the main notes of your knowledge network. One way to do this as well
is to research rabbit so they kind of do it for you which is really nice. And over time then you
will get this big graph like network with all the papers. So following this theme of trying to do
research and learning something new I've also realized that the real challenge is not just
finding the information it's also making sense of it. Connecting the dots and turning these scattered
ideas into one common framework. And I realized this is really quite difficult. And one thing
that quite helps is where today's sponsors comes in and that's Brilliant. So Brilliant is really
designed for this kind of transformative learning. The kind that sticks and can be applied beyond the
app. So their interactive courses in math, science and logic aren't just passive. They also train you
to think the way real scientists do and problem solvers do. Through hands-on exploration, visual
reasoning and through a step-by-step process you will gain mastery over any topic. And what's new
this year is that they've also personalized the experience. So Brilliant adapts to your background
and pace giving you the exact targeted practice and tailored review so that you build real
understanding and just finish a lesson. And one of the courses I really like currently is this course
on scientific thinking. So it focuses down on breaking down problems, testing assumptions and
building reasoning patterns from the ground up. The same skills you need for evaluating resources,
spot patterns and structure your own learning goals. So if you're starting a mini research
project or a mini learning project, setting new goals or just trying to simply build new learning
habits, Brilliant is an easy way to stay consistent and actually make progress every day. To start
learning for free go to brilliant.org, Charlotte Fraser or scan the QR code on the screen or click
the link in the description. Brilliant is also offering 20% off an annual premium subscription
which gives you unlimited access to a library full of courses. So check Brilliant out and let's get
back to the video. So in general you now have this large knowledge web but this is just your starting
point. So you have all these resources from the review but then what you want to do is from those
resources you also want to backtrace citations. So who is citing whom, who is citing where,
and all of this will then build out your knowledge network basically. You also want to forward trace
so if you have a big note or a paper that's cited by a lot of other papers in the field you also
want to see which other papers are citing this paper as well. And one way to do that in Google
Scholar you can actually see which other papers cite your paper but you can also do that in
Research Rabbit for example you can see which other papers are citing your paper. And the thing
is with this literature review is you want to make it really wide so you basically don't want
to miss any of the main papers in the last five years that talk about the question that you're
researching. And as you can see this is already quite an evolved process and you will already
find a lot of resources. So the more specific your question is the less resources you will find and
that's actually a good thing because otherwise this knowledge network will explode. So as you're
doing this and you're going over these papers you also want to clip, save quotes, highlights and
pages to for example Zotero or Notion to really make this second brain idea where you note down
everything that you're learning along this research process. You want to identify themes
that appear across resources and you also want to keep in the back of your mind always what is
missing, where are the gaps, are there certain specific other papers that I should read and go
over such that your knowledge network over time is very complete. So also as you are going into
or through these papers you also want to keep a critical thinking mode hat on and this is a little
bit different from learning. So usually when you learn a new topic you pick one book that is given
by your professor for example and you just blindly trust the knowledge that is in that book. But as
you're doing research or finding papers you actually don't want to blindly trust the resources
you're selecting and the reason for that is that not all research is good but also that a lot of
research is limited. So it might be that they're quite biased or that they're promoting a specific
idea that has not been fully proven yet. So you really always want to be critical when reading
research papers. So something that I do for every paper that I read is I check for biases. So I check
who funded it, is it big pharma, if it's big pharma maybe keep that in the back of your head, does the
tone sound objective, do I think the arguments they present are valid and not from a very pious
point of view, does the data match the headline. So this is something I quite often notice is that
the first the headline the title of the paper doesn't match the data or the introduction the
questions they ask can actually not be answered with the data. So for example sometimes I read
something like depression is caused by chocolate but then when I see the paper that they are citing
I see that paper was actually done only on mice. So the headline should have actually read
depression is caused by chocolate in mice and I think quite often authors or in popular scientific
journals these headlines are tailored to yeah gather a little bit clicks or to make a little
bit of a controversy. But you want to be really aware like what is the data actually telling me
and not just the title. Also something to keep in mind is correlation versus causation. So in most
neuroscientific research we're actually talking about correlational research and not causational
research and the reason for that is that we often cannot manipulate directly human brains because
that's not ethical. So most of the studies we have to do are correlational or we're limited by the
type of interventions that we can do. Also learn some basic fallacies when you're reading papers
that you should automatically flag, how do they deal with outliers, do they use a lot of anecdotes
instead of real research and all of these kind of topics. You will over time kind of get a sense
for this but it's also good to remember that oftentimes when you read the discussion of a
paper it's a very good sign if the research disagrees with itself to a certain degree. So
most researchers know the limitations of their own studies and they will put that in the discussion
for you. So if you read a discussion that's very thoughtful and points out the clear limitations of
the research that's actually a good thing because the reason that there are limitations is because
research is so much at the edge of knowledge and we cannot do all the experiments that need to be
done to answer all the questions. So good research shows these limitations and guides you through the
limitations of their work. So after you have this huge knowledge plan and then your question might
be how do I start reading from this or how do I start learning from this or gleaning information
from this and this is really where you want to create your own research plan or your own learning
plan and you kind of almost can treat this like a personal university course or curriculum where
you basically set out daily time to go over some of the resources and I usually start from the top
notes and then I go to the more peripheral notes and again how you can do this is through Research
Rabbit or if you link yourself all the papers together you will see which ones are the top
notes or the top papers that get cited the most by other papers. So you want to set a daily time
that you will go over these papers and will start reading that and for me that's quite easy because
I do this for work but if you don't do this for work you can for example pick a dedicated place,
time and topic that you will go over and that can for example be before your work or after your work
depending on how much time you have and then something I usually do is to build this self-study
notion template. So I do it in notion again you don't have to do it in notion but you have this
project page, your notes database, a literature review table, this can also be in Zotero, an open
questions list, daily logs of what you learned, what you've done and some summaries of the papers
that you've read. So just taking down a few snippets of what you're actually reading and over time this
self-study notion template really becomes this goldmine of information. So something that I really
want to emphasize here is to try to document everything and I really think your future self
thank you. So basically this notion template becomes like a mini lab book over time where
you record what you read, what you've tried, what field, open questions that you have as you're
researching and try to note down everything and the reason for this is as you're doing research
your thoughts about the research will actually change over time and I think the first questions
that you have when you get into any type of topic are the most interesting because those questions
really show the eyes of a novice and it's quite hard to remember what questions you had when you
research more about a topic. So really try to note down everything you're learning and all the
questions you have and then after you have all this knowledge something that I always say on this
channel is to try to build something from it. So if you have this knowledge bank try to create for
example an essay out of it, a chapter, a video script, a paper or a blog post and the reason for
this is that as you work with the knowledge or you try to answer the question that you started with
I think you really start to consolidate it and just passively learning knowledge or reading
knowledge is not really a good way to truly get into a certain topic. So I would definitely
encourage to write a blog post, you don't have to publish this of course, but to try to shape
the knowledge into a coherent answer on the research question that you started with and this
is really then the idea of active learning versus passive learning. So passive learning for example
is only reading and just highlighting and just scrolling and active is really this retrieval
where you teach what you've learned, you can also try to close your eyes and try to go over the
story that you've learned or you for example try to explain it on a whiteboard and everything you've
done. But I really think it is very important to not confuse passive learning with actual learning
and that is also when you do research I think it's really again like quite good to create
something out of it afterwards so to go from this research to then action and also something that I
try to do at the end of every research session is to end with a few questions that I try to answer
so that is what did I learn today, what makes it unclear and what's my next step and of course I
don't always exactly write it down like that but I do try to take some quick notes of what is the
next step or what do I still want to learn or where is the missing information and the reason
for this is over time this really builds momentum so if every time you sit down every day for to
research this topic and you see from the previous day what is still missing it's quite easy to
follow up on that on the next day and then you create this research or learning momentum over
time and I think this really allows for your curiosity to kind of push you forward and
especially with research which can take one year two years or even five years for a specific topic
over time it can feel quite tedious and there is so much to learn and so much we do not know and
you also really start doubting yourself but I think if you note down all these thoughts and
all of the questions you have over time I think almost the questions become more interesting than
the answers that you find so I really hope that this research process or as I lifted the veil a
bit that you've learned something from this and that you might use this process to guide your
research as well and I'm just really curious what are you researching currently what topics are you
interested in is there anything you would like me to make a video on I would also love to tell a
little bit of my research one time here on YouTube but I always find it really complicated because I
feel that the research I'm doing is so new so I don't really want to talk about it online because
I feel then people think that this is already the truth whereas something with research that
I always have in the back of my mind is that it's still so malleable so maybe what I'm actually
showing today will be disproven in 10 years whereas I think online or also on YouTube it's
very much a platform to show what's truth and what is already out there but I would love to
know what your thoughts are about this do you think there's a good way to show research that
we as scientists are still really unsure about and how do we show this scientific
process in a way that doesn't claim any type of truth or validity I would be really curious
so let me know down in the comments below and otherwise I see you next week bye
