Category: Insight
A Good Map is Hard to Find

A Good Map is Hard to Find

By | | Comments 0 Comment

The idea of mapping information is heavily used and widely favored today.  There are mind maps, geographical terrain maps, all manner of mathematical graphs to map relationships, and maps for “landscape analysis” used to summarize the state of the art in many fields.  But it turns out that when I look around the discovery literature a good map is hard to find.

Clearly I am biased (as evidenced by “Spark Point” and “The Idea Mill”) toward thinking about things in a map-like framework of (1) focusing on key points and connections, and then (2) refining and re-articulating those elements into a nice, neat shareable package.  At that stage, to me, the map becomes an externalized physical model that can be manipulated and played with, letting you toy with the underlying knowledge cluster sketched out by the map.  And going back to “The Physicist’s Repertoire”, if scientific discovery involves both content and skills then one might want at least one map outlining each arena.  So what kind of map might I use?

Mind maps are the easiest choice—free software or pen and paper, associative thinking, unconstrained.  But mind maps are so free form that the permutations are endless, making it hard to assess if adaptations of the map are fruitful; there can be too many options to try.  Luckily, I came across two other maps that seem to me to have more promising bones.

One is called a “territory map” from Susan Hubbuch’s book Writing Research Papers Across the Curriculum.  It lays out central points in a topic, the hierarchy of points, the direction of ideas between points, and the relationship between points.  This may just have been devised as a drafting device, but it strikes me as a potential foundation for a research tool.  If one laid out a set of knowledge, like scientific discovery skills, as Hubbuch suggests then you would have a territory map representing what is known, perceived, or believed.

Then you could play “what if?”  What if a given sub-hierarchy changes, or a directional was reversed, or relationships were added or subtracted?  Now since Hubbuch’s territory map also has built into it a “beginning” and an “end” (again, it’s designed for drafting a paper with an introduction and a conclusion) then that means there is an overall flow from foundation points to supported conclusion.  So, in a skills map, could this flow run from actions taken to supported outcomes?  In other words, could it be fashioned into a draft of a decision-making tool (more usually called a decision tree)?  If so, it could be a powerful way to articulate and refine scientific discovery paths.

Another possible type of map comes from Sanjoy Mahajan’s The Art of Insight in Science and Engineering, in a chapter outlining the technique of using “easy cases” to reduce complexity in order to foster insight.  The author calls it an “easy-cases map” and it’s essentially a flow chart showing the change of a wave equation between ocean regimes and the physical meaning of each regime.  It caught my eye because I once studied the reflection of sound waves, for submarine sonar under various ocean conditions, as part of a high school internship.  And I never felt I actually grasped the relationship between domains of different ocean conditions.  Where was this map 20 years ago?!  Better late than never I guess.

Mahajan’s map-like synthesis, especially between regimes bounded by some key variable or other (which is all-pervasive in physics), strikes me as so potentially useful.  Mahajan’s mathematical map is very much the counterpart to Hubbuch’s conceptual map.  The more variations of either map you have, for the same question or discovery goal, the more you can explore.  Because once something is mapped then you can compare maps for similarities and differences—it’s a powerful multipurpose abstraction.  The key would always be to capture the most “useful” features in a map, so that the meaningful similarities and differences that can act as a spark point for discovery jump out at your perception (which is much faster than cognition).

For now, I have started drafting my first map of discovery strategies and also one of open questions in neutrino physics.  The process will surely be iterative.  But who knows: I may find that the act of mapping and iterating itself will have a part to play in my pursuit of discovery, and in any case, when you’re out pioneering you can never have too many maps.
The Insight Exchange

The Insight Exchange

By | | Comments 0 Comment

I have in mind a way to foster cross-pollination and more interdisciplinary interaction among the sciences (both physical and social).  Initially, I had a hunch that such cross-pollination might be the key to my own discovery efforts as well as to that of others.  This is a little bit obvious in the case of researching scientific discovery itself as a topic, just by looking at the range of fields that publish literature on the process of discovery (psychology, sociology, philosophy, history, literary criticism, linguistics, etc.).  But it is maybe less obvious that this “meeting of the minds” across fields has a place for more technical scientific problems.

But a little reading suggests it might be useful.  J. Rogers Hollingsworth’s sociological study of institutional factors affecting discovery suggests that scientific diversity is important.  Though it is not at all clear to me what “diversity” means to Hollingsworth.  Across the physical sciences—so physicists, and biologists, and chemists, oh my?  Or among physicists–astrophysicists, condensed matter physicists, and medical physicists?  Or even narrower—just within particle physics: neutrino physicists, B meson physicists, and lepton flavor physicists?  I’ll have to dig more into Hollingsworth’s other work to find out.

Another intriguing study I found was a working paper by a group at Harvard, authored by Karim R. Lakhani et al. (I have not yet found out if the paper was ever published in a journal).  They did a study of an unusual situation: a for-profit company that posts open scientific problems, unresolved within the private sector, for a large (300,000+) group of scientists from backgrounds across all the sciences to help solve.  In other words, crowd science.  Scientists choose among the open problems they want to attempt and companies who obtain problem solutions award the solver with financial awards (in most cases).  What was intriguing about the study is that they found a positive correlation between the solver’s perceived distance to the problem field and their likelihood of solving it.  The more “outside” the solver, the more likely they were to solve it.  The devil is in the details though: what precisely do they mean by “’outsiders’ from relatively distant fields”?

But, again, it seems to me that there is room here to facilitate a “meeting of the minds” where there is the shared goal of “how do I discover something new?”.

On considering another article in Harvard Business Review by Greg Satell, describing breakthrough innovation with a quaint example about clams, pollutants, and microchips, I gather that, where there is a well-defined problem, opening it up to “unconventional skill domains”, i.e., other fields with other knowledge and techniques, can lead to simple, powerful solutions.

Hence the idea to start the “Insight Exchange”: facilitated sessions of small groups of truly cross-disciplinary scientists to discuss discovery, sticky problems, and strategies to make progress.  I look forward to hosting a first pilot test Exchange this coming Fall semester with a good-natured bunch of unsuspecting colleagues.  I’ll use much the same 90-minute format as the “Wisdom” session I facilitated as part of the joint University of Melbourne-Vanderbilt University Ethical Leadership Course one-week retreat that I participated in as a physics graduate student.

At the time, how I was finally assigned with such a deep topic as “wisdom” (as a physics graduate student?!) I don’t know.  But I suppose it set a precedent for being a bit fearless in the face of another such deep topic like scientific discovery.  And even more importantly, the assigned reading strongly biased me toward re-envisioning scientific discovery as a skill set that can be taught.  The first reading for the session was by Aristotle (Nichomachean Ethics, Book 6, chs. 5-8) where he stressed that a goal was prudence: the greatest good attainable through human action in a given situation.  The second reading was by a famous American psychologist, Robert J. Sternberg, who said that children should be taught wisdom as part of their school curriculum, as a well-defined skill made up of two teachable parts (having an awareness that things evolve over time and perceiving and acknowledging the legitimacy of opposing viewpoints).

Teach all children how to practice wisdom.  Do not wait and hope that wisdom will somehow find at least a few people, likely in old age.  This was a radical notion to me.  No doubt, my present beliefs about scientific discovery—teach all researchers how to practice scientific discovery, don’t wait and hope it strikes a few lone genius wunderkinds—have grown out of this vision.

With any luck the Insight Exchange will become a valuable real-world source for discovery tactics and diffusion of discovery strategies.  I mean what I say when I say that I believe that a multistream approach is necessary to science: as the water analogy implies, tributaries and streams allow you to cover vastly more ground than sticking to one large mainstream river of thought.  And if you create a venue to allow that multistream to converge in a shared reservoir, pooling resources, then you can truly harness the wisdom of the multistream.
ARTEMIS

ARTEMIS

By | | Comments 0 Comment

For an overview of the ARTEMIS project status, click here to be taken to the ARTEMIS (VR Software) page.

I’m finding the most difficult (and intimidating) part of pursuing discovery to be coming up with new ideas, at least on days when my “systematic mind” is team lead.  On these days I stick to knowns and try to refit combinations of knowns, or incrementally push ideas a little toward the boundary of the unknown parameter space.  A gentle shift here, a nudge there, but nothing really addresses the underlying discovery level shifts needed.  It’s as if I’m pushing the same pebbles around the table expecting an oil painting to appear.

On other days, systematic’s co-director “imaginative mind”, takes hold and I am overwhelmed with ideas, but drowning in the ability to sift and evaluate them.  At those time it’s as if there’s a canvas in front of me wild with splotches of thought, but no clear scale against which to weigh the relative merits or value of each, a splotch at a time.

In essence, it’s impossible to hold these two perspectives in mind at the same time.  But it’s also difficult to capture the outputs of each perspective in a way that lets me glide back and forth between them so that I can make meaningful progress on the crux of pursuing discovery: conceiving of something new and having the prudence to recognize that idea’s significance.  Which sounds to me like it’s time to find a good tool to augment the process.  As the saying goes, is there an app for that?

Certainly, there are mind maps, endless note taking software, pen and paper, LEGOs, clay, foam models, scientific visualization and more.  But none of these are purely designed to foster human conceptualization, let alone human conceptualization about Nature, through the modality of science.  In particular, as I’ve started to read more deeply into research on scientific discovery, and as I think back on my own experiences and difficulties in ideation and follow-through for truly novel ideas, I’m struck by how all the strategies revolve around mental models and reasoning skills; in other words, messy, qualitative, human thought, not structured, quantitative, human calculation.

When I then add to the mix the fact that I know this will need to be able to work with speed and to foster thinking up ideas as much as thinking about ideas and working with ideas …  Well then, I conceive of a tool I’ve nicknamed ARTEMIS, Artificial Reality Tool for the Enhancement and Manipulation of Insight in Science, whose job it is to help you recognize the undiscovered, both in Nature (i.e., scientific discovery) and in your own understanding of Nature (i.e., scientific insight).

I envision this tool will work in virtual reality, where images, sounds, and direct hand manipulations will be the mode of operation; feeding your perceptual and sensory-motor mind as much as your cognition to aid in ideation and evaluation; moving language, computer code, and mathematical symbolics to the background, all of which are slower and cognitively more cumbersome.  Most importantly I imagine it will run in different modes and allow you to enter your own research questions as various abstractions designed to trigger different innate reasoning skills linked to insight and discovery.  And it will allow you to evaluate among options at times when you are a fountain of ideas or to find new streams of thought on days when the well of inspiration runs dry.

I already have in mind a few neutrino questions to use as test cases, to hone, refine, and infuse ARTEMIS with everything I learn in my pursuit of the process of scientific discovery.  The power of virtual reality as a serious research tool remains untapped, especially in its ability to redefine the relationship between humans and computation.

In the act of discovery one can think loosely of three phases: (1) conceptualization, (2) calculation, and (3) interpretation.  (I take here a different and more physics theorist-based view of the phases of discovery than philosophers, psychologists, or historians might.)  Much, much work has been done on automation in physics, and tools abound for precision calculations.  Interpretation is also receiving its due with the advent of numerous tools for scientific visualization.  But conceptualization remains neglected.  Perhaps because it requires augmenting human thought rather than human action.  Whatever the case, I am designing ARTEMIS to fill that gap and serve as a tool on the path to scientific discovery.

And as for the name, perhaps a bit of a lucky chance that I could think up such a quaint acronym; for in pursuit of discovery, who better to have as a traveling companion than the Greek goddess of the hunt.