ReRepost: Something scary - generating big ideas.

Jan 07 2014 Published by under [Education&Careers]

It's been a while since I dragged this 2009 post out of the archive in 2010, but it fits with an interesting conversation that has cropped up at Scicurious' blog. In her post, Sci claims that she should have been kicked out of grad school earlier because she never had good ideas. My issue with her assessment is that she seems to assume that people either have good ideas or they don't. I don't know how other people do things, but for me, it's not a issue of having great ideas but a question of identifying tractable ways to attack a question that you want to answer. In most cases, these efforts DON'T ACTUALLY ANSWER THE QUESTION, but make headway in this pursuit.

IME, this is where junior people (myself included) get stumped. They see a question but no way to answer it and become frustrated or think they aren't smart enough to figure it out. What needs to be learned is how to find work arounds to have short and long term goals that set you on the path of answering a question. Once you have building blocks you might be able to make the jump, but it can't happen without a decent foundation. My lab is STILL in the early stages of chipping away at our central question. But we're making progress and finding some really cool stuff along the way.

Ideas are not a bolt of lightening or something solved in sudden insight at 3:00am. The wear more like a river through rock and occasionally there's an advance that allows you to burst another dam.


With Halloween this weekend, I thought I would post about something that recently scared the crap out of me: Coming up with my own Big Idea.

As a grad student and postdoc, it's essential that you are always coming up with your own ideas, but you have the net of working in a lab with an established theme and having lots of people around working on related things to bounce ideas off of. Then you start applying for jobs and have face the fact that you need to sell yourself on your own ideas. Some people might be able to leave their postdoc labs with projects of their own design are will continue working along those lines. That's great if you can pull it off and it will sure make your life easier. Of course, I didn't do that.

I spent an inordinate amount of time trying to find a way to take advantage of my fairly diverse training in order to come up with a novel research program to pursue, but coming up with an independent and exciting research direction is a daunting task. I had lots of ideas, but either they borrowed heavily from what I was doing at the time (and I didn't want to compete with my PDF advisor in my early career) or I wasn't excited by them. This went on for a couple of weeks. Reading. Thinking. Repeat. It sucked, because I couldn't shake the feeling that I was going to end up either doing research that only slightly excited me and 6 other people in the world, or not doing research at all because no one wants to hire someone with boring ideas.

So, I took a different approach. I started thinking of it like a layered database, where the top layers were huge questions that could not be directly tackled and each successive layer below became more and more tractable from a research standpoint. You can't write a grant proposal saying you want to cure cancer, but you can say that you will use XX cell line to understand YY process with the ultimate goal of making headway towards treatments for a certain type of cancer. My problem was that I was looking at the top and bottom layer and couldn't connect them until I used this approach to think about it.

I started with a broadly-observed phenomenon that I was very familiar with from the work I was doing as a PDF and tried to figure out ways to explain how things transition between the normal and altered state. In order to do that, I decided to look outside the systems that people had used to make the observations and identify a system where the actual transition was ongoing. The search for the right system led me back to my PhD training, where I was introduced to a truly unique system that hadn't been worked on in years. With my question and system in hand, all I needed was methodology to make the observations I needed and do the experiments to test the system, much of which I had learned as a PDF.

In retrospect, it all makes sense but I can't tell you how many hours I spent trying to see how I could carve out my own scientific niche. And hell, I haven't gotten anyone to pay me to pursue these ideas yet, so they might still all be crap. But I do know for a fact that my questions and the unique system I am using to go after them had enough of a "wow factor" to make a big difference during interviews for a job.

That's just my experience, but I doubt I am alone in facing the daunting task of making a research program one's own. It's unbelievably scary to feel like you can't come up with the one original question that you will need to make your mark, but having a broad knowledge base and getting into some of the older literature is what allowed me to piece things together. It's an exciting time when you;re finally on to something that you can turn into a unique research program.

4 responses so far

  • Anon says:

    How many years of work in a field do you think one needs to be able to do this? And by "this" I mean being able to put together a coherent research/career plan to make you a competitive TT candidate at an R1. Is it a bad sign if you can't do this at the end of your PhD? How about after 1-3 yrs of post-doctoral training? When might it be time to throw in the towel if you recognize that you simply can't deliver on something that seems to be non-negotiable for a future PI?

    You write: "As a grad student and postdoc, it's essential that you are always coming up with your own ideas ...." But what's not really clear to me is what you mean by "ideas" here. A grad student/postdoc can be very creative when troubleshooting stuff in the lab -- these are ideas. They can tweak an algorithm or a protocol to answer a "big" question that their advisor wants to pursue -- these are also ideas. They can even contribute an idea or two to a proposal that their advisor is writing. But I think what I'm asking about -- and possibly what Sci was getting at -- is not so much an idea, but a vision. Having a viable vision that you can convince others to let you pursue seems like quite a daunting task to this grad student, who after almost 4 yrs has only made what I consider piecemeal contributions. But such a vision, it seems to me, is a prerequisite for landing a TT job. When did you get to the point of being able to articulate that?

  • proflikesubstance says:

    Part of my point is that it is not as though you just see it all of a sudden, it takes a lot of hashing through. It took me a couple of months of thinking, reading and bouncing ideas off of people before I had a nascent plan to start that I thought was truly original. And by original, I don't mean new invented methodology to ask a question no one had ever dreamed of. I mean pulling lots of pieces together to address a question in a novel way using established methodology. Over the years that plan has developed and evolved into what my lab focuses on today.

    Would I have been able to put our current plan together after my PhD? Not a chance. For me, the integration of what I did as both a graduate student and a postdoc was key to finding a niche.

  • FSGrad says:

    In light of the previous post of the cost of classical training, do you think there is a tradeoff between being classically trained but expected to come up with one's own ideas / dissertation as a student versus being cutting edge but less independent?

    Obviously these don't have to be in opposition, but in my limited experience they have been. Part of that in my subfield is funding: more classical, experimental fieldwork is just less expensive than -omics and therefore the students who do it depend less on their advisors financially and are more intellectually independent as well.

  • pyrope says:

    Maybe using the term 'big' puts too much pressure on junior PIs. I tend to think about things in chunks that consist of either a paper or a proposal/set of related papers. Thinking about doable ideas that advance science seems like a more realistic career move and one that creates less anxiety. I still find myself occasionally looking around and thinking, 'shit, when I finish this paper/proposal/whatevs, I'll be all out of ideas'...and then somehow that doesn't ever happen (yet).
    Echoing your above advice though - Read! This should be totally intuitive for most, but for some reason I spent a fair bit of time thinking that I should be able to pull everything straight out of my head. Turns out my head ain't big enough 🙂

Leave a Reply