Randomly Walking through Research

From reading papers, it’s easy to gain the following picture of what the research process looks like: someone starts at point A, a known point in the space of knowledge, then directly proceeds through various arguments and data to one’s conclusion at the previously unknown point B.  However, thinking that research actually works this way based on what you see in a paper is like thinking that Michael Jordan just awoke one day and suddenly starting dunking from the free throw line.

No, MJ almost certainly traveled a long road to get that much air.  The same goes for research.  The real research process more resembles the famous physics concept of a “random walk” (or more colorfully, the “drunkard’s walk”).  In a random walk, some process is imagined as an object, perhaps an inebriated human, taking a step in a random direction at regular time intervals [1].  This idea is used to model everything from chemical reactions to stock markets.

The random walk provides an interesting visualization of the research process as well.  Uri Alon, a scientist at the Weizmann Institute in Israel (whose outstanding set of resources for “Nurturing Scientists” will be a topic for future posts), has described the process as the following [2].  You indeed start out at A, headed for B.  (See figure below.)  But instead of a nice straight route, you embark on an irregular trajectory with many detours, barriers, and delays.  Often you are eventually forced to abandon B altogether: B was already discovered by some Russian guys in the 1970s, or maybe it’s impossible, or perhaps it just can’t be reached if you hope to graduate within the current decade.Image

At this point your random walk enters a limbo state that Uri Alon calls “the cloud”: you know you can’t go to B anymore, but you don’t know where else to go.  Being stuck in the cloud is probably the most difficult part of doing research.  But the key is recognizing this is a natural and inevitable part of the process.  If you persevere, you will leave the cloud by eventually finding a new place to go: point C.  In fact, often C is much more interesting than B would have been anyway — the unexpected almost always is.  Of course, sometimes C also fails to work out, too, in which case you redirect to D, E, F, etc.  (Hopefully you don’t run out of letters!)  The point is that research is less like a direct path from A to B and more like a random walk with an unknown trajectory and an unknown destination.  But after all, it is this journey into the unknown that makes research so exciting and so important.

[1]  Mlodinow L.  (2008)  The Drunkard’s Walk: How Randomness Rules Our Lives.  Pantheon, New York.
[2]  Alon U.  (2009)  “How To Choose a Good Scientific Problem.”  Molecular Cell 35: 726-728

2 thoughts on “Randomly Walking through Research”

    1. I think I came into grad school with fairly realistic expectations thanks to some good research experiences in college, so I suppose I wasn’t too discouraged by it. (Although when these delays and detours in research occur now, it’s still frustrating!)

      But I do think a lot of folks aren’t adequately prepared for these realities — in science at least, I blame the culture that frowns upon any “touchy-feely” discussions that might be required to prepare people for the inherently human endeavor that is scientific research. Opening up the culture to these discussions is ultimately the goal of folks like Uri Alon (who’s the main source of the ideas expressed in my original post), who has written and talked extensively about this subject (see “Materials for Nurturing Scientists” at http://www.weizmann.ac.il/mcb/UriAlon/).

Leave a Comment

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s