Thoughts about self control

Prompted by a 2014 paper called ‘The evolution of self-control’ by Evan L. MacLean and lots of others from all over the world.

A paper which was about the varying abilities of animals to inhibit learned motor responses, the response, for example, to look under some particular box for a bit of food. This last being a test for the A not B error invented by Jean Piaget, well over half a century ago. See reference 4.

I was intrigued to read in the paper that elephants, despite their large brains, were not very good at this sort of thing, were always making the A not B error.

I also associate to stories about mobilising armies at the time of the first world war. At that time, in mainland Europe, mobilising armies was all about getting very large numbers of armed & disciplined men to the frontier one wanted either to cross or to defend. In either case the idea was to do it before the other lot did, but my present interest is more about how easy or not it was to stop this process once started.

The men, together with their arms, baggage, equipment and horses, were moved mainly by train. Long railway journeys from the interior of the country in question to the frontier in question. Journeys which might be hundreds of miles long and might take days, if not weeks. I leave aside the laying of the additional tracks which might be necessary for this is to be done in an acceptable time. But even without that, such movements were complicated and required large numbers of movement orders, orders which all had to fit together, like a jigsaw, to make a coherent whole, the whole in question being the movement of an army, in one piece, to the frontier. And it clearly saved the staff officers time if all these movement orders had been drafted, at least in part, in advance: one could then just get down the relevant box file from its shelf and have all the stuff in it fed into the telegraph machine. I leave aside the business of being sure that the telegram containing the movement order which arrives at a unit does indeed come from Army Headquarters – a problem which, I think, first manifested itself at the time of the American Civil War. And the related problem of being sure that no-one could read the telegram in transit who should not. Not to mention someone tampering with the box file while the staff were out at polo. Or lunch.

Being picky, one might add that not all the orders are about movements. One might need, for example, to put in an extra order to the coal man for all the extra coal which was going to be burned by all those extra trains. Which triggers the further thought that when warming up orders which might have been put together some time ago, there are going to be glitches. Perhaps the coal merchant mentioned in the orders has gone bust – and too much of this and the chaps on the ground are going to get into a pickle – the lesson being that plans need to be tested from time to time to make sure that they still work.

Leaving that aside, the crux of the matter in hand is that while the army has to move by rail, the orders can move by telegraph, so the army might have been on the move for some days and it might still be possible to call a halt. Being more precise, it is still possible to call a halt until the first units go past the last telegraph office. Until then, one can always get the telegraphist to charge out of his office waving the red flag. Always provided that the units in question did not have the sort of orders which anticipated and overruled such flag waving; the sort of orders which I understand strategic bomber crews would be have been given during transition from the cold war to a hot one. Once they have been given the go signal, they do not stop for anyone. I suppose the idea here is to stop anyone unimportant having second thoughts about nuking some enemy city, second thoughts about killing millions of more or less innocent people.

A catch might be that if one does call a halt in this way, one’s army is left in a bit of a mess. It would, for example, be a lot harder after such a halt to get the army from frontier A to frontier B than it would have been to head off to frontier B in the first place, where the threat now turns out to be. The railway network would be gridlocked with stationary trains, including the ones carrying the food for the chaps out in front. And the army would be a sitting duck if, when suddenly halted in this way in open country, perhaps strung out along country roads, the enemy came crashing through.

I don’t think the staff work at this time, around a hundred years ago now, would have included prior preparation of all the orders needed to countermand a mobilisation which was already under way. Perhaps nowadays, with computers, an appropriately authoritative general could just press the abort button and the computer would crank out all the orders needed to get the army back into its barracks and dépôts in an orderly way.

All this being by way of an analogy with the central nervous system of a person, variations of themes aired last year at references 1 & 2 – and an analogy which highlights the fact that commands do not happen instantaneously, rather they are processes which take time; weeks in the case of an army and a second or so in the case of an animal. Processes which can which can be interrupted and which can go wrong. An analogy which provides, as it were, a magnified view, a stretched out view, of what in a person takes place within a very small compass, both in time and space.

The issue in this case being the ability of a person, or in the case of the paper with which I started today, an animal, to stop a motor action which has been initiated unconsciously, initiated in the interests of speed but perhaps without taking all the relevant information into account. So the mobilisation planned on the basis of last year’s manoeuvres has been kicked off, without waiting any longer for important information which might be coming in from Agent Orange.

The analogy starts to break down when one observes that the central nervous system does not have a slow track for movements and a fast track for movement orders. There is only one set of tracks and they all work at much the same speed. Although that said, consciously controlled movement is much slower than unconscious movement – at least partly because conscious processing is slow, with things happening at sub-second speed at best, certainly not split-second – with the idea of many training regimes being to make the desired movements more or less unconscious, and so fast. So soldiers are trained to shoot, not to think.

I shift now to the trunk of an elephant, which despite the doubts and difficulties expressed at reference 3, does contain a large number of muscles. So translating the executive command ‘pick up the apple’ into the thousands of carefully sequenced commands needed to move the trunk in the required way is quite a business, which, even supposing that what we are doing amounts in some large part to retrieving a stored program rather than writing one from scratch and allowing for the large brain of the elephant, might take hundreds of milliseconds. The elephant may even have started to issue the first commands, perhaps readiness commands for the first muscles in the program, before it has finished computing the last commands. However, during this time, the elephant might spot something much better looking than the apple, in some quite different direction. At which point the elephant is able to stop, to abort the pick up the apple command, and perhaps after a bit of trunk twitching and perhaps after a short pause, to move onto picking up the chocolate bar.

So in some sense or other there must be more than one set of tracks, more than one route. Perhaps it is in part a volume thing: the many commands needed to make the trunk move can only move along at a fairly stately pace, but the short, sharp command needed to put a stop to the whole business can whizz along the hard shoulder and catch the many up at some junction, some choke point which it is easy to shut off. Thus accounting for the pause while the now unwanted commands dissipate – somewhere or other.

A wheeze which can only be made to work in animals which are conscious? Not that the consciousness is strictly necessary, one could program a robot up which could manage such stuff without, but in real animals, consciousness does seem to come with the necessary access to command & control functions.

Another angle might be chemical preparation for action; that is to say it is not enough to send the command by neuron, one also needs to organise energy supplies by dilating blood vessels, perhaps also to get the heart pumping a bit faster, and one may need to organise supplies of neurotransmitters at various places, at the synapses involved in relaying the commands to the muscle of, in this case, the trunk - harking back here to the purchase orders for coal mentioned above, orders which might or might not be fulfilled by the civilian contractors when they get back from lunch. This sort of chemical action sounds quite time consuming and it may be necessary to kick it off on a speculative basis, before one has decided what exactly it is that one is going to do. It may also be quite broad brush in that while one can send a command by neuron at a particular time to a particular point in the body, chemical action is going to cut a wider swathe. And is going to take longer to undo, if that is what one ends up doing; there may well be chemical debris to flush out of the system. In which connection, I offer as a closing (but unconfirmed) factlet, the waste product content of tears, with the tears serving, in part, to flush out unwanted chemicals produced inside heads at times of heightened emotion.

All in all, a reasonable and informative exercise, or at least it might be if one worked a bit harder at it. One analyses the workings of a visible system in order to inform one about the workings of an invisible one. On the understanding that the invisible one must, in one way or another, at some level, be doing the same sort of thing as the visible one.

PS: much of the military colour in the foregoing has been taken from the Parrott translation of ‘The Good Soldier Švejk’, written shortly after the first world war.

Reference 1: http://psmv2.blogspot.co.uk/2015/10/on-colonels.html.

Reference 2: http://psmv2.blogspot.co.uk/2015/09/an-analogy.html.

Reference 3: http://psmv2.blogspot.co.uk/2015/03/senior-to-elephantine-moments.html.

Reference 4: https://en.wikipedia.org/wiki/A-not-B_error.