Some folks got quite exercised about the “cliffhanger” ending of my post, last week.
This week I will discuss what the original target of the thought experiment actually was, how the experiment morphed into what was ultimately posted, and how I think the utility of the thought experiment has been improved both for the purpose of illustrating my original target, and for use in other areas.
The first thing to notice about the experiment was the reductionist quality. We start with two visually similar “simple” objects that exhibit different properties, but which nevertheless have many similarities, beyond just their spherical shape. Originally, my intention was to allow you, the reader, to use your own imagery to provide one such spherical object – and indeed that is still in the post. However, I decided to provide the polar opposite images of bubbles and ball bearings, not least for the pleasing alliteration.
As with the scientific method, the reductionism in this thought experiment allows us to get rid of the complex features of real-world examples, allowing us to focus on the essential features of the target objects. We can then add the other features back in, and see what happens.
A quick restatement
To begin with, we looked at what would happen in the case of many millions of such objects, crammed together on a flat plane (again, attempting to keep things simple). Then we looked at what would happen to such a collection of “identical” objects if they were to be impacted on one of their leading edges. Of course, immediately we notice that the objects cannot be truly identical as they all inhabit slightly different locations in space, and some are on the outside edge, whilst others are in from the edge.
The intention was to keep these spheres uniform, but the moment you introduce location, you introduce some degree of identity. Furthermore, unless we completely ignore physics (which we absolutely can do in a thought experiment) there will be different types of pressure being exerted on these spheres, depending upon their location on the group. Those on the outer edge experience the least pressure from the group, but are the most open to incursion from outside.
Very quickly, with minimal additional complexity, we have rebuilt concepts of identity, context, and structure. The idea of incursion from outside the group – whether from foreign objects or from clusters of former members of the same group – highlighted further the ideas of identity, context and structure. In addition, the idea of incursion by former members of the group showed the interaction between group membership and contextual complexity. In other words, the fact of an additional individual entity alters the properties of other such entities, the addition of more entities eventually leads to more groups, and both alter the environment in which they exist.
The Original Target
My original target for this thought experiment was the human brain. I wanted to illustrate the logical structure that underpins thought, both rational and irrational, with the intent to do away with the trope about how intelligence cannot arise from matter. The spherical objects, with whatever properties you dreamt up, were going to stand in for neurons that we could get to behave in fairly uniform ways, at least to begin with. So, what I will do now is develop the thought experiment in the intended direction, and then bring it back around to explaining just why it might be so useful in highlighting interactions and behaviours in other, erm, spheres.
Somewhere between a bubble and a ball bearing is a spherical object which is more solid than a bubble, but with more give than a ball bearing. Let’s go with a baseball (sticking with the alliteration already established). So lets add this third spherical object to our ontology, in order to talk about brains.
Image from: Psychology Today
The bubbles, ball bearings, and baseballs, exhibit particular behaviours when collected together in a single homogenous group on a flat plane. In all cases we end up with some kind of matrix. In the case of bubbles, because they have a flexible outer surface, a honeycomb-like structure will form. In the case of the ball bearings, because they don’t as readily deform, we get a tightly packed collection, depending upon the amount of pressure that is maintaining the group perimeter. A baseball will deform a little bit, on for or six sides, depending upon whether the rows of balls are offset, and they will not deform as much as a bubble and, like a ball bearing, the baseball is unlikely to burst.
As mentioned previously, individual bubbles are more likely to break if something impacts them, depending upon the viscosity of the film that encapsulates the gaseous interior. There is some transference of the energy of the impact radiated out to surrounding bubbles, but it is quickly dampened (especially if the impact is from other bubbles). By contrast, in the case of the mass of ball bearings, the energy of the impact radiates out, potentially traveling all the way through the mass, with the possibility of bearings coming away from the mass on opposing side(s) to the impact (like a Newton’s cradle). With baseballs, such an impact would radiate out in fairly predictable ways, similar to the ball bearings, and the energy being transmitted would be dampened as the energy passed from ball to ball, much like bubbles.
With a matrix of near-identical spheres, what happens when it is impacted at several points at the same time?
When it comes to the brain, there are a fairly circumscribed number of entry points. Which is to say, the impacts or incursions mentioned so far have been completely random in nature whereas, for a brain, such incursions come only from a few sources, and the signals from those impacts dissipate across the mass in a similarly circumscribed set of ways. Recall that I mentioned that two impacts near to each other could reinforce or cancel each other’s signals. This is analogous to excitatory and inhibitory nerve signals. If particular patterns of signals predominate, then the mass is going to orient in a way to dissipate that energy across the mass as efficiently as possible. Balls may stick together more through repeated impaction, and equally, balls that are not in the path of energy dissipation may lose their stickiness to nearby balls that are using that stickiness as a resource for survival.
We have location, and thus identity, context, and structure, and we have tapped into this further by having signals only entering the mass through certain points. It may be that structures, having become optimized for these directional “signals”, as described above, but such a system may also come to process correlational information. This is to say that a structure that is suboptimal for either of two signals, but mostly suitable for either of two signals that are often repeated, will do better than a structure that is optimised for one, and not the other. This could be used to explain why it’s hard to not see faces when there are two points either side of vertical line formed by the combination of a somewhat vertical line, and the centre of a horizontal line. Face detection is optimized, by correlation, our ability to consider the elements of the face separately, especially once we’ve seen a face is, vecause of that, sub-optimal.
Fundamentally, almost anything can be described in this simplified (but not actually simplistic) way. It highlights the interactions between similar entities, but equally it highlights just how much of the behaviour at the complex level is predicated on “behaviour” at a simple level. I could as easily describe the action of water flowing over stone in terms of simple spheres that, while softer, are greater in number and velocity, and that stone will therefore erode, and the water will optimize it’s path accordingly. I can describe the actions within a cell, then within the life form that contains that cell, then the planet on which that life form resides.
The important thing to note is that bubbles are made up of smaller bubbles… or maybe smaller ball bearings! As such, I’m not standing on the edge of an infinite regress – we know the levels at which we can describe entities. For example, these simplified spheroids can be engaged to discuss things at the molecular level, then the cellular, then per organ, and then per organism, and so on up. Indeed, a number of papers have described the fuzzy boundaries of social groups in terms of the permeability of the cell-wall (Dechesne, Janssen & van Knippenberg, 2000[i]; Ellemers, Spears & Doosje, 1997[ii]), and even extended the concept to a “sociocultural homeostasis” (Damasio, 2010[iii]).
As I mentioned at the start of the original post, this is but the start of an attempt to build a thought experiment that allows us to talk in simple terms about complex things (as such, I wouldn’t use th word viscosity when trying to use this as an xplanatory tool for children). Primary amongst my aims was to be able to talk about the interactions between these “identical” spheroids in the terms of folk physics, thereby making an explanation grounded in this idea accessible to almost anyone. The intent being to make it a useful, not actual “lie-to-children”, but an appropriate level of simplification for all laypeople of any given (though probably mostly scientific) subject.
[i] Dechesne, M., Janssen, J., & van Knippenberg, A. (2000). Derogation and distancing as terror management strategies: The moderating role of need for closure and permeability of group boundaries. Journal of Personality and Social Psychology, 79(6), 923.
[ii] Ellemers, N., Spears, R., & Doosje, B. (1997). Sticking together or falling apart: In-group identification as a psychological determinant of group commitment versus individual mobility. Journal of Personality and Social Psychology, 72(3), 617.
[iii] Damasio, A. (2010). Self comes to mind: Constructing the conscious brain. Random House LLC.