Organizing science: A closer (albeit ancient) look at the field of cognitive psychology
In a recent blog post I promised to dust off an old informatics diagram I developed eons ago. nSCI is asking informatics fans around the country to send in their ideas, so here’s an example.
First, let me offer my apologies:
- This is old, pre-computer era work so some of you out there will recognize the look of papers that were cut and pasted together and graphics that were drawn by hand. It wasn’t pretty.
- I’ve forgotten much more than I can remember, and this informatics diagram (and the accompanying paper that explains it) is a great example. I couldn’t defend this work today if my life depended on it. I’m not going to try, though, since doing so would obscure the point here. This work is dusted off for your review and entertainment to illustrate what it might look like when people in the know (i.e., not me) really put on their thinking caps and try to visualize science in broader terms.
- This isn’t dissertation-level work, just graduate work, and looking at it now I can see that it’s not properly cited and the reading level is about 19th grade — in other words, it’s not something I would put my name on today but it’s still worth looking at for reasons explained below.
So let’s get on with the dusting-off. The impetus for this work was that long ago — back in the mid-1980s — I had the good fortune of taking some graduate courses in cognitive psychology with one of the field’s most prominent scientists, Dr. Elizabeth Loftus (who is now an nSCI board member). I was intrigued by how much wasn’t really understood about the field at the time — especially coming out of a science undergrad environment where knowledge stood on firmer ground. And I was fascinated by the prospect of organizing what psychologists did know. It seemed to me at the time that organizing this knowledge would help lead to more discovery — that the field seemed to be hunting and pecking for disjointed pieces of discovery, and that if these individual pieces of knowledge were organized, a bigger picture of psychology might emerge. Psychology is not alone in this hunting and pecking, of course — all science looks like this, which is why the potential for organizing knowledge might be so promising.
The organizational model that I developed was pretty simple — essentially, looking at a few hundred studies and then identifying which ones were related and which were basically subsets of broader studies. Then, these studies were grouped into a hierarchy — a knowledge tree. You can view the whole tree by clicking here; the paper that went along with this tree is here.
You really need to be a student of cognitive psychology to understand everything that this informatics diagram and its accompanying paper talk about, and this blog post isn’t the best place to go into that kind of detail — it would complicate the basic point. And this basic point is that diagramming science can lead to insight. Never mind that your diagram might be utterly wrong (which mine probably was and is) — it doesn’t need to be definitive, exhaustive, and accepted. In the case of this model, organizing the field of cognitive psychology into a hierarchy revealed where the gaps were in research (in this model, concepts known at the time are capitalized and connected with solid lines; hypothesized relationships are in lower case and connected with dotted lines).
This model also grouped phenomenon in such a way that the answers to larger questions became apparent. For instance, the concepts of “perception” and “interference” — describing how certain influences shape or interfere with how memories are formed — were well-established at the time. But researchers didn’t formally distinguish between perception frameworks which were permanent in nature and interference frameworks which were temporary or transient. By organizing the available research data into a hierarchy, it became evident that perception phenomenon could grouped into different categories, and that these different categories might have different underlying psychological and physiological foundations and consequences.
This model also suggested pathways to new discovery, highlighting how a little understood function called “saturation” might play a critical role in brain function. Did the brain actually contain a “saturation center” and if so, how did it work? This model contained a companion description about how the saturation center behaved precisely like a capacitor, and that the principles of capacitive overload, discharge, and so on described the kinds of behaviors one might expect to see as a result of overstimulation, lack of attention, and more. Neuroscience has changed so much since the 1980s that this companion idea barely rates mentioning now; there is no doubt lots of research in this area that has established more credible ideas and approaches, but the point is that at the time, this area of research was wide open.
Finally, this knowledge tree of cognitive psychology also revealed how perception preceded cognition (which was a hotly debated topic back then), and how much influence our expectations and interpretations of believability and likability have in how we react to and interact with the world.
As was noted in this paper at the time, this model wasn’t intended to be exhaustive — it was just a first approximation for standardizing terminology and the hierarchical significance of perception-related concepts used in the researching of human perception.
So in summary, this approach — if it were carried out today by real scientists who then shared their ideas with peers — might result in:
- identifying potentially new areas of research
- clarifying relationships between existing areas of research
- focusing research on areas of critical importance
- revealing new ways to look at and explain groups of related information
- revealing entirely new phenomenon, and
- bolstering understanding of poorly understood phenomenon.
Does this approach to knowledge organization have any scientific merit? Probably not. But it does deserve a place in science. For me, this is the first, best definition of informatics — not computer-based knowledge extraction systems, but human-based knowledge visualization systems. There are a wide variety of ways that knowledge can be organized, so picking just one might be arbitrary and scientifically suspect. But our brains can be amazing organizers of information — we can often intuit patterns and relationships better than supercomputers — so it’s important to recognize that there is definitely a place in science for intuition. In any event, right or wrong, the insights we gain from our guesswork and experimentation might lead to other insights and guesses that will help narrow down our focus on the right answers. And these guesses and attempts to create order from chaos can help us develop a greater understanding of what is, what isn’t, and what might be.