15

Okay, so I am a math student currently on leave from a graduate program in math and computer science.

I have always been good at coursework. I ace examinations and manage good grades. I can understand and explain concepts well. All my professors have liked me because of all this.

But somehow it feels like I have tricked my way up. Everything goes wrong as soon as I take up a project or open problem or any kind of research. I start out beautifully- read and understand all necessary background instantly, get up to date on current results, impress my guides with my clear understanding. But things go down fast after that. I lose motivation, I get lost in digressions and extensive learning. Or I just slack and avoid the project altogether and somehow try to make a dignified exit.

This has happened too many times now. Each time I thought it'll all come together the next time. But nothing has changed. I have tomes and tomes of notes and expository material on all topics remotely related to my research areas. But I don't have a single original piece of research. It got so distressing that I was diagnosed with depression, OCD, etc. and subjected to medication and therapy. But somehow the problem seems beyond all that.

I really want to do research and solve problems. Even an insignificant, but original, result might greatly elevate my confidence. Are there others who have been through this? Is this just a difficult phase some people go through? Or am I just an idiot who's reaching beyond his level?

14

It is hard to give a good answer here, but it seems to me that you, despite being a good student, haven't learned to do research yet.

When you describe what you are good at, you describe all the things you need to be a good student. These things are all necessary to be good at research, but alas, they are not sufficient. To do research, you need to shift to a different gear.

For example: Do not read a paper thinking "How does this work?" but more "What does that mean? What can I do with it? What could be done differently?". Learn to read articles and books through different lenses, e.g. with a lens that focuses on broad understanding of concepts, a lens that focuses on fine details, a lens that focuses on open problems or possible improvements,…

Also, find a problem to work on, not a topic to learn. Ask some supervisor to provide a problem to work on at the beginning, but note that later on you should come up with interesting problems on your own.

  • True. I realize that myself. I have always taken comfort in the luxury of established theory as opposed to venturing into the unknown. You get the satisfaction of having learnt something without the uncertainty and lack of structure that come with original research. I am not sure if thats just plain intellectual laziness, or a latent fear/low confidence, or simply a limitation of my intellect. "Do not read a paper thinking "How does this work?" but more "What does that mean? What can I do with it? What could be done differently?" Thats the root of the problem. – BharatRam Jan 4 '17 at 17:36
  • 1
    Do not read a paper thinking "How does this work?" but more "What does that mean? What can I do with it? What could be done differently?" — Don't forget "Is this actually correct? And even if it's correct, is it actually good?" – JeffE May 5 '17 at 16:46
7

Notice the language you're using to describe yourself and your behavior. This is most likely the same kind of language you're using to talk to yourself as well.

The highlights in bold are mine.

I have always been good at coursework. I ace examinations and manage good grades. I can understand and explain concepts well. All my professors have liked me because of all this.

These are absolute claims. "always" and "all" are absolute quantifiers. They're great for things like logic, math, or computer science, but they tend to distort our thoughts when it comes to interpreting our own human behavior.

But somehow it feels like I have tricked my way up. Everything goes wrong as soon as I take up a project or open problem or any kind of research.

I start out beautifully- read and understand all necessary background instantly, get up to date on current results, impress my guides with my clear understanding.

But things go down fast after that. I lose motivation, I get lost in digressions and extensive learning. Or I just slack and avoid the project altogether and somehow try to make a dignified exit.

"things go down after that"

Here, it might be more helpful if you were more specific.

Is it as soon as someone stops overseeing your work that you lose motivation? What kind of structure did your department/advisor provide for you - if any?

This has happened too many times now.

How many times did it happen really?

Each time I thought it'll all come together the next time. But nothing has changed.

Really, nothing has changed? Are you sure about that? Please notice the narrative you're telling yourself. Is this claim even true? Many times, we just don't notice all the growth we've made as a student (until years later).

I have tomes and tomes of notes and expository material on all topics remotely related to my research areas. But I don't have a single original piece of research.

"I don't have a single original piece of research." Many graduate students don't either.

Besides, I bet you can find some original ideas in your own work. My bet is that you're just not able to see them right now.

It got so distressing that I was diagnosed with depression, OCD, etc. and subjected to medication and therapy. But somehow the problem seems beyond all that.

For someone who is depressed, this is a perfectly normal feeling to have.

Does your therapy already include a Behavioral Cognitive Therapy component? If not, I'd suggest you take a look at BCT. Also, I'd suggest you listen to audiobooks or videos by Byron Katie, which is another take on the same underlying ideas. She has a lot of free materials on youtube and on file-sharing networks.

  • Thanks for the insights Stephan. I agree that there are more absolutes than necessary in my descriptions But I wanted to convey the main problem without being too verbose and nuanced. The broad idea is captured best by the absolutes, and it wasn't my intention to oversell my virtues nor undermine myself too much.. I did try CBT but am not sure if it was formal and complete. I'll have a look at Byron Katie. Thanks again. Always glad to hear ideas. – BharatRam Jan 4 '17 at 17:32
4

Other answers give many excellent specific tips, but just to pick up one point that hasn’t been addressed yet,

Are there others who have been through this? Is this just a difficult phase some people go through?

Yes, it is absolutely a phase that many beginning researchers go through.

The transition from coursework to research is a difficult one. I remember feeling absolutely the same worries you describe for most of my first couple of years of grad school, and I’ve heard multiple other mathematicians (some very successful) talking about having felt the same way. Finding this a large and difficult step doesn’t mean you’re an idiot who isn’t cut out for research. This is a big new kind of skill you need to develop, and different people find it comes more or less quickly or naturally; so don’t be surprised or disheartened if it does take a while for you, just keep working at it, in the ways suggested in other answers.

  • 1
    Thats encouraging. Though every time I see my friends and colleagues making some progress (however insignificant), I can practically feel my heart sinking. I don't think its envy, but just agony and self-pity over my staying-at-the-same-place. Sometimes I question whether I am throwing good money after bad by clinging on to a futile hope that I'll be able to break that ceiling. Feels like living out the gambler's ruin example. The only thing keeping me intact is my love for the subject and desire to be surrounded by it- even if I am the stupidest person in the room. – BharatRam Jan 4 '17 at 17:45
2

One big difference between coursework and research which is worth remembering -- lecturers won't set you coursework problems which aren't possible, but many research problems are (at least where we currently are).

If you try to solve if P=NP (for example), you won't get far. If you are tackling a genuinely hard problem, it can take months to make significant progress. I will often start by nibbling off little sub-problems, and prove some very trivial results, to approach the main problem. Research ideas which are worth publishing are often months of work, of which at least 80% of which was a waste of time (but of course, you don't know in advance which 80%).

In short -- you might just be expecting research to be something where you make progress every day, and in a fairly linear fashion, which in many research areas it isn't.

  • +1 — expectations from coursework are a big part of the problem. – Blaisorblade Jun 2 '17 at 17:13
2

Are you too much of a perfectionist?

I wanted to make an album for 10 years. I had all the songs written. I had a band. We played the songs. But we couldn't record anything no matter how we tried. Eventually we broke up.

My standards were too high for someone who never recorded an album.

Years later (when I heard Axl Rose was finally releasing his album after 14 years) I said "That's it, I am recording my album with what I have in the room. It will be the output of the next 2 weeks, whatever it comes out as".

Basically, I dropped all expectations and ambitions that made it so hard to record. But I got it done. And then I recorded another one, and another one, and another one. Each one got closer to my original ideal. But I had to come up with a format to get things done, which is a completely different skill altogether.

To write new songs, I would go to a coffee shop and say I'm not going home until I have a song written. I got new songs written after that.

When recording, I would say I have 4 hours to record this. That's 1 hour per instrument. If a part isn't perfect after 1 hour, I move onto the next. I could always go back and tweak it, but at least I had a finished product in 4 hours.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.