I just started my PhD and this is my fourth week. I understand that as a PhD student, my main aim is to find a gap in the knowledge and then trace this gap by using scientific approaches that I might choose later.

My question is: How can I find this gap?

I can think of two options:

  1. Reading, which is the ideal option and we all know that. However, it might be frustrating to read a lot because maybe you will get lost in the final stage due to information overload.

  2. Just jump to the conclusion of any paper in your field and see if they provide any recommendations about a potential gap and choose it as your hypothesis or the starting point for formulating your question.

  • 49
    Seems like the sort of thing that Ph.D. advisors are for. As a beginning grad student you are not expected to have a good knowledge of where the research frontier is. Commented Nov 6, 2017 at 16:13
  • 5
    @JohnColeman, there are some Ph.D advisors though, who actually the only thing that they say (and is generally true) is: " is your personal research and you are the leader of it". Reading I think is really important on this stage (I am also a new PhD student).
    – geo_dd
    Commented Nov 6, 2017 at 16:22
  • 5
    Perhaps reading some review papers on your research area might help!
    – The Guy
    Commented Nov 6, 2017 at 17:15
  • 14
    @geo_dd It is probably somewhat field specific. When I showed up to grad school in mathematics, almost all of the math that I actually knew was discovered before I was born (and often by a century or more). It never would have occurred to me that I was supposed to figure out as a first year grad student what I intended to do my dissertation research on. I didn't have the background to understand what the relevant knowledge gaps were about. Commented Nov 6, 2017 at 17:17
  • 15
    is your personal research and you are the leader of it — Sure, but there's a difference between "take the lead" and "Here's the deep end; I hope you like sharks! [shove]".
    – JeffE
    Commented Nov 6, 2017 at 20:37

8 Answers 8


I would not suggest starting by looking for gaps at all for a few reasons.

Some gaps are left gaps for a reason. Some are not relevant enough to be bothered with, some have not enough data to work with, some lack technology to work with. In any case, even if you find a gap, the follow up question will be why do you want to spend time and resources trying to fill in that gap? That is a pretty hard question to answer.

Instead, try to start by finding a problem. Look for a problem that is relevant, that is worth solving, that is even conceptually solvable. And the most important thing, look for a problem that you are passionate about because you are about to spend a good chunk of your life trying to solve it and will probably fail for the most part. So every bit of success you have should be meaningful for you and for others who are affected by this problem.

Once you found the problem you want to help solve, then you can look for gaps within that narrow area. That should be a piece of cake because chances are you will bump right into them and later find yourself surrounded by them.

  • 5
    I like the warning that gaps may be gaps for a reason. But the rest of the answer makes me wonder what the difference between a problem and a research gap is. I think there's a valuable observation somewhere, but it needs some fleshing out. Commented Nov 7, 2017 at 12:44
  • 2
    A gap concerning a gaps.
    – Pysis
    Commented Nov 7, 2017 at 14:06
  • 5
    @Abra Sure it is. Read paragraphs 3 and 4. The answer is "find it later (implicitly using either/both of methods 1 and 2) after narrowing to a specific problem".
    – Dawn
    Commented Nov 7, 2017 at 18:28
  • "later find yourself surrounded by them" this is very very true (at least in my subject)
    – YYY
    Commented Nov 8, 2017 at 16:37
  1. Ask the advisor.

  2. If he/she is helpful, follow their advice.

  3. If he/she is not helpful, either

    • change the advisor (which is preferred) or else

    • find an interesting and open problem in your field (in some areas such as mathematics, searching for "open problem" or similar in recent publications, including literature reviews, might help). The problem should be such that that you should be able to solve it within the time allocated for your PhD studies, including writing your dissertation. Then, ask the advisor again whether the problem is suitable for your thesis. How to proceed from there would be a different question.

  • 14
    I wish it were enough to look for the words "open problem" in recent papers to find interesting but approachable problems in math... Commented Nov 6, 2017 at 18:25
  • 5
    I too would beware a bit of the last hint. When somebody puts an open problem into a paper, it often (but of course not always) means that they tried to solve it for a while and did not even see where to start. Otherwise they would have done it themselves or given the problem to one of their students. Considering the difference in experience, you are likely to fail, even if you put in more time then they did.
    – mlk
    Commented Nov 6, 2017 at 20:36
  • @mlk but then the thought process of another human brain might give an insight into a problem that the former brain(s) could not, even when the former brain whas the same but from some time ago. Doesn't happen very often, but there is a story about a very famous bath a philosopher took in antique greece ..... Commented Nov 7, 2017 at 12:54

As someone coming out the other side, I'm not going to give you two answers. The easy answer is to just ask your advisor and have them give you a project. I've seen plenty of people try this and the problem is that if it doesn't really grab your interest then grad school will be a grueling frustrating boring experience.

The alternative is the far better solution, at least in my opinion. That is simply, don't look for it. As counter-intuitive as that sounds. Most programs don't expect you to have your project figured out the first semester or even the first year in many cases. Take that time to really dig into the parts of your field that interest you. Really dig deep and learn all you can learn. As you learn you'll have questions. If you're doing it right, they'll be the kind of questions that keep you up at night. You really need those answers to satisfy your mind. Go look for those answers. When you come to the question that is vexing you and you can't find the answer and nobody around knows where it is, then you've found your gap in the knowledge. Only this time it will consume your mind and you will be able to think of nothing else. And grad school will feel like this beautiful opportunity to have all of these resources behind you while you try to answer your question. It will almost be fun. It will be rewarding. And it will keep your interest. And when you get done, you won't just have the degree, but you'll also have all the other stuff that can come with it when you really do it right.

  • I wish I could double upvote this answer. By becoming an expert in your field, you will gain a good understanding of where there are significant gaps. Commented Nov 7, 2017 at 14:03
  • I agree with "Really dig deep and learn all you can learn." Another thing to keep in mind, while doing that digging and learning, is that there are often alternative ways to view some material, ways that others might not have noticed yet. I've found that following such an alternative approach to known material usually leads to a clearer understanding and often leads to new ideas that can become a good research project. Commented Nov 8, 2017 at 2:50

Reading available literature in your field is the best place to start finding what has not been researched in the recommendation portion of research dissertations.This should point you in finding a gap.

  • 1
    You should expand this answer as it currently looks more as a comment. Commented Nov 6, 2017 at 17:20
  • 6
    In my field (chemistry) there is so much literature even on very small subtopics that it would be basically impossible for a new grad student to find "not researched" but still interesting and relevant topics within a reasonable time.
    – user64845
    Commented Nov 6, 2017 at 18:15

There is some good advice in the earlier answers. To find the most suitable one you must know yourself of course: For some people it works to not focus too much (as user82540 suggests: just read and let your curiosity do the magic), others can be given a question and take it from there all alone (the supervisor must have the topic all sorted out), while others need something introductory to get their hands on before they get any sense of orientation (often that's an assignment at grad school). It is ideal if you have all options, but usually one doesn't.

Therefore, beyond the already good answers (including your own), I have the following to add, which may be especially useful if your university has no comprehensive grad course in your field or if your supervisor is not very helpful (and in the understanding that it is not always easy to change supervisors or universities): Attend a relevant summer school (or winter school) early in your studies and don't hesitate to ask the lecturers the same question, i.e., about hot topics in their fields. This is not to say that you will get specific topics from them necessarily (they have their own students), but you are bound to get some direction. Be prepared to 1) first show them what you have been doing (anything: masters thesis, what you read that interested you...) 2) sit back and hear them brainstorm (some are exceptionally good at that) and 3) understand maybe 1/10 of it.

Such schools often offer hands-on experience too (e.g., in physics you get to use somebody's computer program and see science at work) and for many people that's very illuminating.


Something which worked quite well for me (in maths, that is), was combining several (sub)topics of your field. At the beginning of your PhD might be the last moment where you have enough freedom (in terms of time) to look at different areas you know nothing about. So try to gain some breadth of knowledge.

Even by the end of your PhD, you will probably not know as much as the experts on topic A or on topic B. However you can be one of the few with both good knowledge about A and about B and thus the only one to realise, how to apply some method from A to something from B in a novel way.

In other words, instead of only looking at the cutting edge, the gap of knowledge you need might just be a gap between two subareas.

  • 1
    This is good general advice about life. You'll never be the best in the world at X, but it is very possible that you have a combination of W, X, Y, and Z skills that nobody can exceed. Commented Nov 7, 2017 at 0:25

if you feel your advisor is not helpful, then change advisor.

The advisor should tell you at least the area to look for (the topic) as well as a sound hypothesis (that may not be extremely fine tuned but can be a starting point). Very rarely the advisor will actually give you a discovery.

Your role as a PhD student is to make a discovery, this will happen when you will put the results of your experiments (made to address the initial hypothesis) in the context of the current knowledge (which you have to acquire by reading, going to seminar, ask more experience people, etc...)


This might bespecific to your discipline, but I think you do have a misconception about what your aim as a PhD student should be. The aim is to do research and provide new insights into your field based on scientific methods.

To me, the fact that you have started a PhD position without a topic is a bad sign. Your supervisor should either have given you some ideas or asked you what you want to do research on when you applied. Imo if neither of those has happened you should seriously reconsider whether he is suitable supervisor to your (or anyone else).

Now given that you are in this position anyway and you might not agree with my opinion here, I would give you the following advice.

  1. If you are searching through literature for problems to solve, try finding out if review papers are common in your discipline. Those usually are very long papers which give a great overview over a field once there has been substantial progress and if there are known experts around. These are to my experience more likely to make you aware of open questions and unsolved problems than other types of papers.

  2. If you are working in social sciences it might also be helpful to search your own head. What I mean is: Why are you interested in this field? Which open questions do you have? What did people fail to explain or prove to you during your studies?

  3. Again for literature review I would focus on more recent problems and publications. Often you will find that you can use methods old methods and apply them to a new field which they have not been explored in. This gives you a quick start into a topic and once you start doing research on your own more interesting questions might pop up.

If you have done all this and figured out a general area of interest, go talk to your supervisor. You should do this either way. Again he does not necessarily have to come up with a very explicit suggestion, but anything that gets you to start working on a topic where he - or some of his close colleagues - have a good level of expertise might open exciting questions for you to work on along the way.

The same could also happen if you start collaborating on something with another PhD student or postdoc. I really advise against pure literature review for the search. Personally I couldnt keep that up for more than 2 weeks straight.

Not the answer you're looking for? Browse other questions tagged .