Thursday, October 16, 2014

What makes a scientist creative?

Science is about generating knowledge, but it’s also about the process of generating knowledge, and few things delight as much as creative ways to generate knowledge. Some of my favorite examples include ribosome profiling from Jonathan Weissman’s lab, or Michael Elowitz’s two color noise experiments. Not that all scientific progress comes from creative experiments, nor do the results of all creative experiments stand the test of time. It’s just that these are the ones that are so awesome that you never forget about them.

Some scientists are just really good at coming up with creative ideas (Sanjay Tyagi, my former PhD advisor, is one of them). Where does scientific creativity come from? There is I think some notion that creativity is an innate ability, but I’ve come to think of creativity as a skill, which has an important distinction: skills can be learned and honed, whereas innate abilities cannot. Some amount of creativity is innate (perhaps having as much to do with interest in a topic as raw brainpower), but if you have someone with the raw materials to be a creative scientist, then you can help shape that material to make that scientist more creative than they would be otherwise. How? Does some of this just rub off from the mentor to the mentee? What in particular is it that can rub off?

I’m guessing there’s a lot of psychology research in this area, but here is a thought that I had recently. It came from an e-mail I had with one of my (very creative) trainees, which was an awesome moment as an advisor. I had just e-mailed the trainee, posing a question like “hmm, what are the implications of these results.” My trainee wrote back, saying “well, could inform x or y”, which is pretty much the current thinking in the field. And then I got another e-mail 10 minutes later saying “These are both silly answers. It is definitely something to continue to think about.” I was so proud!

This exchange got me thinking that maybe one of the underappreciated elements of being creative is just not settling for being not creative. If you are in science, there’s a pretty good chance that you have ideas, probably many ideas, maybe all the time. The key is really in the evaluation. When am I just settling for the status quo of thinking? When is the status quo probably right and there’s maybe nothing here? When have I really hit the foundation of the problem we’re working on? If I could do any experiment to test this, possible or impossible, what would it tell me? What is the closest I can approximate that in the lab? These are all things that we can consciously think about and that mentors can teach their mentees, and I think it can help us to be creative. I also think that establishing a rigorous culture of idea generation and evaluation can help the group as a whole become more creative.

Thinking about creativity reminds me about when I was in a band back in college. The leader of our band, Miguel, was one of the most creative people I have ever met–lyrics and music came out of him in ways that seemed mysterious and divine. (Incidentally, I feel like not settling was a big part for him as well.) He was really good friends with this other amazing songwriter named Joel, and Miguel used to say “You know how I know that Joel is a better songwriter than I am? Whenever I play someone a song I wrote, they say ‘Man, how did you ever think of that?’ When they hear a song Joel wrote, they say ‘Oh man, why didn’t I think of that?’” Same applies in science, I think.


  1. YES! Glad to see you are teaching mentees how to push boundaries. Sometimes I think the pursuit of creativity gets superseded by practical constraints, like available energy and resources. It's still something I highly value and hope to emphasize to students of my own.

  2. Edward de Bono is the best author on this theme, I think

  3. I tend to lean towards the "creativity is an acquired skill" school of thought. From personal experience as someone with a lifetime of experience as a carpenter/builder, I can tell you that my problem solving skills in this domain are highly correlated with my experience having dealt with thousands of problems in this domain in the past. As our skill set increases, the thought tools we have to come up with creative solutions increase as well.

    This is a subject studied in depth by sociologist Richard Sennett in his book The Craftsman (Yale University Press 2008) He states:

    "The difference between brute imitation of procedure and the larger understanding of how to use what one knows is ... a mark of all skill development." (p58)
    "... as skill progresses, it becomes more problem-attuned, like the lab technician worrying about procedure, whereas people with primitive levels of skill struggle more exclusively on getting things to work. At its higher reaches, technique is no longer a mechanical activity; people can feel fully and think deeply [about] what they are doing once they do it well." (p20)
    "In the higher states of skill, there is a constant interplay between tacit knowledge and self-conscious awareness, the tacit knowledge serving as an anchor, the explicit awareness serving as critique and corrective." (p50)

    Sennett gives an example:

    "In the Linux network, when people squash one "bug", they frequently see new possibilities open up for the use of the code. The code is constantly evolving, not a finished and fixed object. There is in Linux a nearly instant relation between problem solving and problem [opportunity] finding."

    This seems to point to a way of widening the creative pipe; develop an open not closed knowledge acquisition environment where the process of execution and technique critique/technique refinement go hand in hand.

    - Markus F. Robinson