Tuesday, January 06, 2009

Factory of Ideas

Before I got distracted by the awesome Statement of Purpose Contest, I'd started drafting a post about the generation of ideas for research, inspired by a recent Chronicle of Higher Education essay by Robert Hampel ("In Search of New Frontiers: How Scholars Generate Ideas"). Zuska beat me to it with an interesting post on this topic, but I have a few things to add.

I think it is interesting and useful to discuss strategies for generating research ideas and what exactly constitutes a new idea for scholarly investigation and thought. The Chronicle article contains good advice about how you can cultivate some intriguing ideas for scholarly pursuits (e.g., have conversations with colleagues at conferences, over lunch, in a cafe, wherever).

Hampel mentions several colleagues, each of whom has a different method of generating and/or organizing ideas. The variety of approaches is apparently evidence for a lack of "coaching" in graduate school or during other educational experiences. Hampel wrote:

Surprisingly, no one I spoke with had been taught how to generate topics for future research during their years in graduate school. Several said that since research had pervaded the ethos of their university, they had merely absorbed the spirit of curiosity.

Why is it surprising, and why is it a problem not to have been taught in an organized way about idea generation? Would it be a good thing if everyone had the same method of idea generation?

How do you teach someone to have ideas, other than by example? Isn't absorbing the spirit of curiosity a major step towards generating ideas? As a student, you don't have to be told explicitly by a professor "OK, now I am going to teach you how to generate ideas". In grad school, you learn by doing, you learn by watching, you learn by absorbing, and then you figure out how you want to do things.

I definitely think it is good to have conversations with students and postdocs about some of the idea-generating concepts discussed by Hampel and Zuska (and her commenters), and I think that another important role for faculty and other advisors is to give students and postdocs the confidence to express and develop their own ideas.

In the course of our advising and teaching, we can provide information that helps our students and postdocs to develop ideas and recognize what is a good idea and what might not be such a good idea. As advisors, we also teach others how to follow through on an idea. That is, once you have an idea or a glimmer of one, what do you do about it?

Even if your grad project and/or postdoctoral research involves doing tasks related to someone else's ideas, by doing research, you are gaining skills, either specific ones or general ones, that should be useful when you are in a position to work on your own ideas. By reading papers, listening to talks, doing research, and other basic activities of the academic life, you should be able to acquire sufficient knowledge and experience to make a start on your own career of ideas.

I also think that students can take some initiative and ask questions about these things. If you really have no idea how researchers generate and frame an idea, you can initiate discussions with various people in your academic environment. Most of us advisors probably assume that our grad students are learning these things as they go along and asking questions as they occur.

When I was a young assistant professor, one of my first Ph.D. students* came to talk to me. He told me that he had some general questions about being a professor, and he wanted to discuss these so that he was well prepared for the day when he was a professor himself. I told him that I thought this was great and that I was happy to discuss these things with him. He paused, looked around my office for a moment, then asked "How do you decide what labels to put on your filing cabinets?".

There are some things you can teach.. and some things you can't.

* who subsequently failed his exams and went to another university and then dropped out of that school and I have no idea where he is now but I hope he found something that he enjoyed and was good at doing


Anonymous said...

That shit with file cabinet labels is fucking hilarious! And yet it is also sad that a grad student could be so utterly clueless about where the ball is; it's not at all surprising that he failed to become a scientist. I see this kind of thing sometimes when I interview post-doc or faculty job candidates: you can just immediately tell based on questions they ask--and ones they don't--that they aren't even in the right building.

In relation to idea generation, inspired by Zuska, I posted my own thoughts about this at DoucheMonkey:


This excerpt describes my conclusion concerning the suggestions in the Chronicle article: These sources are certainly reasonable, but subject to a fatal flaw. They are each guaranteed to only lead to future research directions that are already visible from where one currently stands. Different approaches are required to generate truly novel--that is, unexpected and unpredictable--ideas.

Anonymous said...

Hey! Filing is a serious concern.

When I was a grad student in a large shared office we each had a cabinet and our own ways to file papers and there were many arguments over which method was best -- alphabetical by first author, organized by research project, alphabetical by journal!? [crazy Germans!], etc. All of that is moot now that we have tags and electronic papers, but, still.

However, if that is your biggest question about being a professor then maybe you should really be going to library school instead.

Prof-like Substance said...

I think this is an important aspect of science culture, but like you say, not something that can be taught the same way bench techniques are. Plus, arriving at novel concepts may happen in entirely different ways in different fields. I posted a bit about this as well and I think ideas in my field are developed (for the most part) in a different way than, for instance, CPP's field. Nevertheless, teaching how to have ideas is like teaching how to have curiosity.

Dr. Jekyll and Mrs. Hyde said...

Actually I spent the better part of second year grappling with how to design a filing system that worked for me. It's hard! That said, I wouldn't have put it under "career development."

Anonymous said...

I definitely think it is good to have conversations with students and postdocs about some of the idea-generating concepts discussed by Hampel and Zuska (and her commenters), and I think that another important role for faculty and other advisors is to give students and postdocs the confidence to express and develop their own ideas.
What you suggestion is an ideal situation at best. My postdoctoral advisor is very judgmental so having such a conversation with him would be extremely difficult. Plus I think he is hurting for new ideas. I have lots of new ideas. When I first came here, I suggested combining my new expertise with my former expertise and then he wanted to do it.
I think that in order to be creative, one has to give up being judgmental. Creativity flows best among people who are not negative. The most judgmental people lack creativity. Just an observation.

Anonymous said...

I take a lecture in my grad courses to tell students about different ways one can go about first selecting a problem to work on, and second start to break it into manageable pieces.

However, as with almost any craft, that is just the theoretical component and can only be truly learned by them doing it. The second part of the course consists on the students selecting a problem to work on, with ample feedback from the instructor.

The hardest part to learn is the judgment of when a problem is ripe for solving and no less importantly when it is worth solving.

Global Girl said...

Although I agree that the idea generation itself is difficult to teach, vetting of ideas certainly can. I also think that knowing how to vet ideas can help in coming up with new ones - if you know the general characteristics of a good (executable and interesting) research idea, it must be easier to recognize one. Perhaps it is really this that it meant by "teaching innovation."

Project management is very relevant here - a project is a temporary undertaking to achieve a unique result. That describes pretty much all research. While project management knowledge and know-how kicks in primarily after you already have the project idea, thinking about things like the final paper/thesis chapter/book chapter/whatever (the deliverable in project management-speak) can certainly flesh out an inkling of an idea, because it forces you to think wider about a *set* of results you'd like to get or questions that need to be answered. The inkling I think curiosity will bring you, as you say in your post.

I know there are advisors and groups that teach thinking about the paper at the idea stage and then using the paper to guide what you do (like Whiteside's group). Project management goes even further in the same direction and can help you consider things like how much time it will take and whether you really can do what you want to do before you start. Those practical restrictions have helped me think out of the box and also to think of new ideas as my evaluation of inklings wanders along.

I am a certified Project Management Professional, but my project management knowledge came from involvement with a student-run consulting firm rather than my research environment. I have thought many times that knowing what ways have proven better or worse to run a project would be very helpful in research overall, but there is an acceptance barrier because project management is seen as too "businessy" and not academic enough. I disagree sharply, and think that teaching project management basics is as close to teaching how to come up with new ideas one can reasonably get.

Indre Zl said...

I spent 3 months in Bangor University, Wales last year and they had a bunch of graduate special purpose courses. I was able to attend only two and one of those was creative thinking. This was a half day course where they tough PhD student the techniques how to generate ideas. And the practice followed. Another course was on preparing posters for conferences, not limiting to the contents, but also the colours, formatting and a talk on printing services provided by the university. I haven't attended, but I heard they also had a course on selecting the conferences where to submit the papers and submission process for PhD students.

Anonymous said...

I think that in order to be creative, one has to give up being judgmental. Creativity flows best among people who are not negative. The most judgmental people lack creativity. Just an observation.

This is absolutely true. There is not single motherfucking new idea that cannot be rationally shot down. Creativity comes from having the fucking balls to try shit that you are sure won't work. Half of my lab is now pursuing a vast new area of inquiry that was opened up by a technique that we invented that I was 99% sure wouldn't work before we attempted it. We tried anyway.

Anonymous said...

A few months ago I was trying to decide whether to apply as an M.S. or Ph.D. student. Since my main worry about going for a Ph.D. was the ability to generate my own novel ideas, I asked my dad (also a professor in the physical sciences) how he came up with research topics.

He told me there were two necessary components: (1) read the literature, and (2) ask the right questions about the research. Like, was the new methodology the researchers were using a valid way to look at a problem? If so, could it be applied to problems he'd tackled in the past? If not, what were the issues with their methodology and could they be easily rectified?

Reading the literature was important, he told me, because creativity thrives on exposure to new ideas. If you don't hear about others' ideas, you tend to focus more and more exclusively on your current way of thinking, and your ability to generate novel ideas is sharply limited.

In the six months since we had this chat, I've generated enough ideas for research that I could get through a Ph.D. program and a postdoc. Needless to say, I decided to apply to Ph.D. programs.

Wombat said...

Idea generation is important to open up new reseach avenues. Some of my best ideas come from reading outside my area and attending meetings outside (or ones that peripherally overlap) my area. The most productive project came from reading a popular account of tissue regeneration based on a BBC series.

My first post-doc, although not as spectacular as I had hoped, was with a free-thinking and slightly unhinged PI. Exposure to him gave me confidence to pursue the occasional wacky idea. I have never looked back. However, too many ideas leads to paralysis and a lack of focus.

Of course, you also need the resources to pursue some of these ideas, which are hard to come by these days......

Anonymous said...

About letting ideas flower. It is important to keep in mind that the first approach suggested rarely works as is and pointing out one of the (easily fixable) ways in which it fails is a waste of time.

If on the other hand someone finds an insurmountable obstacle then by all means inform the group so we don't wast time down a dead end.

Nathaniel @ project management course said...
This comment has been removed by a blog administrator.