What should I do when my PhD adviser sends me a paper to read?

Question

I'm currently in a PhD program in Electrical Engineering. Sometimes, my adviser will send me papers with a small note such as "I thought this may be relevant to your work. Please have a look and see if it is helpful." Because I'm relatively new to the field that I'm working in, reading a paper, even if not as rigorously, takes me some time. The main reason is that I am often not familiar with the technical terms in the paper, so I need to look them up while skimming.

While I appreciate my adviser's consideration for sending me paper X, it will often be the case that I am working on implementing something from paper Y when they send me paper X to read. Paper Y could have been a previous paper that my adviser recommended, or it could be one that I found.

Because these papers are technical in nature, so that only reading the abstract would not give me a good enough idea of whether they are helpful or not, I feel that my progress towards paper Y would be slowed down considerably if I split my time between reading paper X and working on paper Y, as the repeated context switching would waste some time. This is especially true if my adviser sends me more than one paper to look through.

I've added below some ways I thought of proceeding, and I'd appreciate any feedback on them, or any another way that I haven't mentioned.

1. Thank my adviser for sending over the papers, and let them know that I will get to them as soon as I'm done with the current paper that I'm reading/working on. I'm not sure if this approach would be considered rude or not.
2. Drop the current paper that I'm reading/working on and focus on the paper(s) that my adviser sent over. While this may please my adviser, I feel that my own progress would slow down, especially since I would be constantly reading over papers that my adviser sends over and not actually getting any work done.
3. Split my time between the papers that I'm working on and the papers that my adviser sends me. As I mentioned, because of the repeated context switching, I feel that this approach would slow me down.

Answer 1

As you mentioned in the comment section, since your supervisor gives you only 1-2 papers per week to read, I think it is totally possible for you to handle more than this. It seems that you spend too much time for a paper, but it is understandable as you are new to the field. It will take you some time to be familiar with the field. Eventually your understanding will be better, but you will still have to improve/practice reading skills. Knowing what to read and how to read effectively is important. Your time is limited.

Also, if your supervisor sends you something to read, it is usually something useful. It does not mean that you have to read it now.

I have my own strategy for reading papers. It does not work for everyone, but it is similar to (3). Assume that I have many papers to read. I choose one of them and spend some time on reading the first 2-3 important sections of that paper. Then I move to another one and repeat the same process. I keep doing it again and again until everything is finished. It also helps me cope with getting bored with reading a very long paper.

Last but not least, it is helpful to take notes as you read.

Answer 2

You may talk about this with your advisor, but I strongly suspect that an advisor sending you a paper does by no way mean that you should read it immediately. Most probably the advisor was doing some research on their own, found that paper, thought that it may be useful for you, and immediately sent it to you so that neither them nor you will forget about it. But it should not be considered an immediate call-to-action.

So what you can do is schedule some time for reading papers, say each Monday morning, and you postpone any new paper till next Monday (unless there are strong reasons to read the paper earlier, of course). (Obviously the period may be different, depending on your situation and suggestions from your advisor.) If before that your advisor asks whether you have read the paper, you simply reply that you were working on <whatever it was>, and you plan to read the papers on Monday. (And obviously do it, and this includes having some system so that you do not forget which papers you advisor has sent to you. This may be a simple "mark-it-unread-in-email-inbox", or a todo list, or just "print-it-out-and-put-into-a-physical-to-read-folder", etc.)

This way you will not have that much context switching, and at the same time you will thoroughly read the papers.

Probably you don't even need to reply to your advisor email, or just reply with a simple "received, thank you" (this depends on your advisor and general culture, of course).

Answer 3

Basically, if they send you a paper, saying it is relevant, then you should read it. If they don't send a paper copy then print it out and put the date received on it so that you can sort them as needed. (See below for "why" paper.)

But you are busy, so you need a way to proceed to make this reading "easy" to accomplish.

Here is a pro tip for any scholar. Always have something immediately available to read no matter where you are. This was early advice to all of our doctoral students.

Whenever you are "out and about" have one or two of the papers with you. Also carry a pen so that you can annotate them, and perhaps a highlighter. Optionally, carry a few index cards as well so that you can take notes external to the papers and arrange them as needed later.

Now, whenever you have a few moments of dead time, perhaps waiting for your lunch to be delivered or you are waiting for a bus (or riding one), pull out one of the papers (or a few of the index cards) that you are carrying and read or review what is before you. This turns "wasted" time into productive time.

Yes, you should thank them. No, you shouldn't interrupt all other work to read the new paper unless the supervisor strongly recommends it.

Answer 4

When I was starting out in my PhD I used to have this same problem that you're having. I realised after a year or so that when my supervisor read a paper, what he meant by this was that he had read the abstract, and then opened it up and flicked through the equations. What reading a paper meant to me at this point was that I would go through every line in detail and try to understand everything I could, following references for things I didn't understand and deriving equations.

So I would advise you that, an important part of the process of doing a PhD is learning how to read texts at different depths. I guess at the start you would be likely to need to read things in a bit more depth than your supervisor to understand as much as they do, but it's unlikely that they expect you to read everything that they're suggesting a lot of detail.

If you're unsure, you could also try asking your supervisor, "how in-depth do you think I should read this paper?", or "how much time do you think it would be worth investing on this paper?"

Answer 5

(This is an attempt to summarize the advice given in the other answers and comments in a more hands-on form.)

Your supervisor is asking you to have a look at the paper, which is not the same thing as reading it. Your supervisor is probably expecting you to quickly browse through the paper to form a first impression of its relevance to your work.

What I'd do is what others are suggesting: spend 15-30 minutes with the paper, and with Google, to answer the following questions:

• What does the paper try to do?
• How can this be relevant to your work?
• Do you think studying it more will be worth your time?
• Why? / Why not?

The answers should all be one or two sentences. Write them down. If you can't decide on an answer, write down which questions you need an answer to first. File the results in your list of known papers. Discuss these results at the next meeting with your advisor; they may be interested in your findings and they should be able to advise you how to proceed.

The goal here is not to understand papers in depth, but to navigate the space of papers. Your advisor can help you with this, but only if you can establish a dialogue on what you're finding.

Answer 6

Welcome to being an academic.

One important distinction that many younger academics don't make is "being ready to hear/read" vs "being ready to understand". Not every paper that is sent to you should be read in full detail. Or even in partial detail.

There is one broader lesson here, the way mentors/advisers interact with you is as much the lesson as whatever technical content they tell you.

When I got to grad school my adviser would talk about math in ways that sounded unhinged and weird to me. But he is an excellent mathematician. It was the way I was talking about mathematics that was wrong not the way he talked about mathematics. He was (still is) better than me. When I defended my PhD we went out to dinner and I was talking in the same "unhinged" way. It's just a matter of information throughput.

One thing that I think is missing from your calculations here and from some other answers is sending papers to each other is how academics interact. Your adviser is sending you papers partly so you can read them and learn from them. They are also sending you papers because that is how you should interact with other academics you work with. I send papers to people literally everyday almost. I get papers sent to me almost literally everyday. This is just being an academic.

As for how much detail etc. do you read them? If you're asking this question, you're thinking about this very wrongly. You're an academic now. It's your decision. How relevant or interesting is the paper? Not every paper should be read the same way. Some I've poured over and others I've skimmed. It's your call because you're the researcher.

Answer 7

Lots of good advice* but I'd like to emphasize that the strategies urged upon you (keep up the physical work, but make regular time for reading papers in addition) will be valuable to you. How many of us find things when writing the thesis introduction and background at the end of the PhD project that we wish we had known before we did the experiments? One here!

*which I can't comment on due to insufficient reputation - else I would just add a short comment.

Answer 8

tl;dr: Talk to your advisor about this.

What should I do when my advisor tells me to [etc. etc.]?

The answer to this question template is most usually: Talk to your advisor and coordinate their expectation from you. Not every time you get this message from them, but when it happens the first time, schedule a chat at their office or remotely; tell them you want to attend to the request/suggestion they made; explain that you are not sure exactly what they expect in terms of scope/breadth/etc., and you would like for "meta-guidance" on that.

It is ok not to know how to best handle such requests and to ask to be told what to do. You will not appear foolish, argumentative or ungrateful.

Specifically in your case,

Because I'm relatively new to the field that I'm working in, reading a paper, even if not as rigorously, takes me some time. The main reason is etc. etc.

Tell your advisor that. Ask them whether they think you should invest the time to read rigorously, or whether they mean skimming the paper, or reading abstract / abstract & intro.

current paper that I'm reading/working on

talk to your advisor about prioritizing or scheduling ongoing work (including reading) with their incoming suggestions.

Answer 9

Authors include abstracts for a reason, my advice is to go through the abstracts, and then prioritize into 3 buckets (read the whole dang paper ensure I understand everything, skim the paper for important points, skip/read later). It sounds like not everything your advisor is sending is going to be 100% relevant (just, maybe relevant). If you do this you will get a sense of which papers are seminal real quick because everyone will be citing those papers. If one of those landed in your skip pile, you pivot to read the whole dang thing.

Now, I know you said that utilizing the abstracts is probably not good enough. I actually agree with that point. You are a student, you need to get your hands dirty with the technical aspects. However, you can still prioritize based on the abstracts. My 3 categories could be broken down like this:

Paper appears directly applicable to your research topic: read the whole paper
Paper appears to only be tangentially related to research topic: skim the paper
Paper appears not to be very relevant for your topic: read later

I would suggest doing your reading with a study journal. I like the hard bound, lined journals, with page numbers and a table of contents section. That way you can create a list of what you read in the table of contents and where you can find your notes on it later by page number in your journal. The hard bound physical journal works well for me because it is sturdy, doesn't require power, and I won't get distracted by browsing the internet. But you do you, it works for me. Some folks like bibliographic tools and they are really helpful when you actually go to write so take my suggestion with a grain of salt.

The other thing is that I wouldn't try to reproduce the papers. There is value in that as a grad student, but you should just try to understand the papers. Make notes about them, and do your own thing later.