Taking on a risky project

[loginofpscl]

My adviser has given me the freedom to pick my project starting out (first-year), and I applied to the school thinking I'd be working on a particularly difficult problem that no one has ever managed to solve before. This problem, if solved, would be an incredible discovery with high rates of reward.

 

There is a current graduate student (~5th year) who is currently working on the same idea-- and has made some very interesting findings and developed incredibly impressive methods to validate the work-- but has yet to publish anything on her topic, and neither will she publishing anytime soon. What do you guys think about taking on high risk, high reward projects with hardly any precedent? I'm pretty intimidated and unsure of how to proceed. Any advice would be appreciated.

[GeoDUDE!]

Often times, when approaching problems like these, you can still publish incremental progress if you do not fully solve it. If that is what you are interested in, go for it. Never let a difficult problem stop you from doing what you want to do. 

[TakeruK]

My advice is to have a backup plan (e.g. side project) in case this high risk, high reward project doesn't pan out. These projects are not considered "high risk" for nothing! I chose to try one such project in my first year and also had another project that was also interesting but less risky (our program has us trying two projects in year one). Actually, after starting both, I found myself much more invested and interested in the second project.

 

In the end, my "high risk high reward" project didn't work out (basically, someone else found the solution first). But it was still a good experience overall because I learned a lot and I am making good progress on my main project. Don't let a difficult problem discourage you but you need to also think of your future career--you don't want to sink many years of grad school into a project that results in very little rewards. In reality, more established researchers generally have more resources and more ability to absorb risk so they are better equipped to tackle high risk high reward type projects. So, my advice would be for grad students to take on these projects as side projects rather than hoping they will be the lucky one to make it happen.

[St Andrews Lynx]

It honestly depends what motivates you. If you feel that you can cope with a long slog of a project (several years of dead ends, false hopes, obstacles, large number of screening reactions) that doesn't yield results but still require a high level of your motivation...then sure, go for it. I think that if you only work on ~3-6 month projects part-time as a undergrad then it doesn't mentally prepare you for grad school projects and how tough they can be. 

 

TakeruK's suggestion is one I completely agree with. Another option is to start out with a smaller, easier project, then progress onto the high-risk-reward project once that one is done. For instance, you could help out a more senior student wrap up their project, or complete a "loose end" that the PI wants tied up. That was my approach, at least.

[loginofpscl]

Great advice. My PI is floating a few ideas I'm pretty excited about. I will ask him about such a parallel approach!