The concept of process used here is a bit softer than that in a conventional computer program, the sort of thing one might write in Visual Basic or Fortran. But while, to coin a phrase, the sands are always shifting, the place known as Dogger Bank is always there. So the population of neurons implementing some process might come and go over time. But it always occupies pretty much the same space in the brain, involves much the same networks and does much the same things – even though the details might change a bit.
I think it reasonable to talk about processes and data in the terms which follow.
I believe that modularity is good and that it is unlikely that the brain would achieve what it does without it, at least in some form or another.
I allow parallel processing. Processes can co-exist and sometimes communicate with each other. In some contexts, this is called multi-threading – with some systems – such as those supporting the use of bank cards – allowing for very large numbers of threads. We suppose that, in the case of the brain, it will be helpful to talk of tens if not hundreds of processes, bearing in mind that any such process structure is being superimposed on a rather softer substrate than that of the chips in a PC, even than that of the thousands of chips in thousands of PCs. This despite systems of this latter size starting to exhibit the same sort of unpredictable behaviour as people.
I also think it reasonable to use the terms id, ego and super-ego. We might no longer agree with the details of the Freudian project, but I still believe in some of the essentials. See, for example, reference 5 for a previous take on some of this.
Super process: id
The workhorse of the brain, running all the time, doing all manner of things and from time to time passing stuff along to the ego for its perusal. Sometimes with some indication of the importance and urgency of the matter, giving the ego some indication of how much time and trouble to take.
Sometimes also, passing along powerful feelings and emotions, feelings and emotions which provide an important backdrop to the activities of the ego.
Super process: ego
Running some of the time, responding to stuff which arrives from the id. More or less suspended when the host is asleep.
Note that we suppose that the id already has a mixture of bottom-up and top-down processes generating its product. The ego is bringing something else again to the party.
We are not here concerned with whatever it is that generates the subjective experience, not here needed.
We started off with the idea that the ego was a very simple choice function, without much access to data, other than that about the options. But it is easy for this to spread out, for there to be mission creep, and for this modest routine to become a brain within a brain, the dreaded homunculus. So what is the right compromise between the ego either being too small to be of interest or too big?
The next idea was of how some company boards work, with, for each board meeting, a pack being assembled by the support staff according to some fairly rigid timetable and given to board members so many days or hours before the (say) fortnightly board meeting. These packs are supposed to be pretty much self-contained and the idea is that most of the time there is enough information in the packs for the board to take the decisions needed without calling for more information, presentations or whatever, all of which would burn up more time, time which the company may not have. Such meetings will be ruled by a chairman who may, nevertheless, be ruled himself by a constitution of some sort. Think ‘point of order, Mr. General Secretary’ in the meetings of trade unions, or ‘point of order, Mr. Speaker’ in the House of Commons. So what could the id deliver to the ego which amounts to such a pack?
For the moment we are running with the idea of a choice model, as set out below. The idea is that we have a kind of duet between the id and the ego, a duet which delivers a choice, an action. But a model that allows a fair bit of latitude as to their relative size and importance.
Super process: super-ego
Running most of the time.
In analysis, I think it is fair to say that the idea is that, during free association, the super-ego should be more or less turned off. That the analysand is reporting reasonably directly from the unconscious.
Here, one might include the idea of cleaning the data up a bit, stripping it down to its essentials. Not cluttering up the very important minds of the very important members of the board with silly and unimportant details.
Data: public memory
Public memory is the data which the ego has to go on.
This data can be made conscious, but only a small fraction of it is going to be conscious at any one time.
Data: private memory
A variety of data supporting the working of the system as a whole, but not available to ego processes. Not available to consciousness.
Data: outside aids
The ego may not have access to all the data in the brain but it does have access to outside aids. Paper and pencil, books and the like. Stuff which can be used to add value to the choice model supplied by the id.
Data: choice model
When the id wants some help from the ego, it will build a choice model which it then shares with the ego. There is a conversation between them, with a view to agreeing on what to do, to agreeing on the choice, on the action to be taken – remembering here that most of the time it has to be just one action. One cannot be in two places at once, so choice can be important. In this, the balance of power between the three components, the id, the ego and the super-ego will vary across time and vary across people. Some people, for example, will have stronger egos than others.
This choice model, very much the sort of thing which might be used by a buyer in a large organisation, supports the choice between the null option – leaving things as they are – and one or more action options, actions which responds to the threat or opportunity presented by the current scene. See reference 6.
Maybe all this data will be held in the form of meta-stable states of meta-stable processes, a thought prompted by the Fingelkurts twins (see reference 3). And maybe these meta-stable processes will be synchronised to their work by gamma waves, or something of that sort, a thought prompted by Buzsáki (see reference 4).
Process: housekeeping
Public memory is decayed. Something which goes on in the background. There are lots of different ways of doing this, subject to the constraint of limited space and the tendency for older stuff to go first. And sometimes it is cleared, perhaps because the id has started on a new scene, a new set of choices.
Process: scene
Start a new round of activity by pushing a new scene into consciousness, into working memory. This is the world as it is, the null option. Something which the id will do from time to time, perhaps as often as every second or so. Depending on the amount of change this may include clearing away any options which might have been there with the old scene, clearing the rest of working memory.
Process: censor
The super-ego, when it is up and running, is given the opportunity to vet content intended for working memory, having the choice of pass, fail or modify. With modify meaning that the proposed content might become acceptable if some of the objectionable content is modified or removed. So for example, a bad image of an indifferent acquaintance might be substituted for a bad image of the important husband. Or a tiresome disease might be substituted for a fatal disease. Sometimes this will be helpful in damping down the ego response.
In so far as option choice is involved, such modification may be unhelpful in that the ego’s image of the world is less real than it might otherwise be. The ego is responding to something other than the real world, which may well result in error or worse.
Is this modification to be done by the super-ego itself, or does the super-ego prompt the id so to do? Following the example of respond in the next but one section, we might plump for the latter.
Process: choice
The id passes along a choice model and then asks the ego to respond to it. The id will go on doing this until it agrees to an option for action with the ego. In the jargon of a computer program, a ‘do until’.
Against the absence of response, there is a wait time, after which the id will move on, without waiting for agreement.
Process: respond
The ego indicates what it thinks about the choice model, in a sense, second guessing the id, using the information provided by the id, together with such relevant information as may be available to it in public memory and outside aids.
The ego, on the whole, can only attend to a row or a column of the choice model at any one time, although it can do better with outside aids. Each time that the ego’s attention moves to a new row or a new column, and from time to time otherwise, the id may take the opportunity to update that row or column with its latest information, for better or for worse.
Put another way and as suggested above, I see this part of the system as a sort of dance between the ego and the id, a dance during which we have the idea of a currency. Lifting the notion from Excel and its worksheets (aka spreadsheets), at any one time we have a current column (an option), a current row (a feature), or a current cell (a score). In the function descriptions which follow, I sometimes fill in the details of the interaction between the id and the ego, sometimes not.
One might see this dance as a conflict or a competition, with the two parties vying for supremacy – while I see it, in least in health, as a reasonably cooperative business where, implicitly, each process recognises the need for the other. A business which is more symmetrical than that suggested by the phrases ‘Her Majesty’s Government’ and ‘Her Majesty’s Loyal Opposition’.
‘Yes’ means that the id can move onto action with the current option.
‘Modify option’ means that the current option has promise and the id is invited to tweak it a bit. This might happen in a couple of ways. The id might have some idea about modifying the current option and pops the command ‘Modify option’ into consciousness, hoping that the ego gives assent. Or it might actually put the modified option into consciousness, then the command ‘Modify option’, once again hoping that the ego gives assent. What the assent does is write the suggestion to memory, a memory which will persist for at least a while, to which the ego has access and to which it will hold the id – with the id on its own tending to be a bit fickle, a bit wayward, with stuff coming and going in a rather disorganised way. And by breaking the process down in an orderly way into steps small enough for the ego to be able to trust the response of the id, the ego can trust and sign up to the result of the process as a whole.
‘Modify feature’ means that the current feature is still of interest but the id is invited to tweak it a bit.
‘Modify score’ invites the id to tweak the current score a bit.
‘Discard option’ instructs the id to drop the current option. Perhaps, for example, because the ego deems it to be unrealistic. Then go back to the option which was offered before the current one.
‘Discard feature’ instructs the id to drop the current feature. Perhaps, for example, because the ego has decided that it is no longer interested in it. Then go back to the feature which was offered before the current one.
‘Beginning’ goes back to the first option, the current scene. Which may have moved on a bit since the ego last looked at it. The tiger may have got a bit closer.
‘Top’ goes back to the first feature.
‘Last option’ goes back to the option which was offered before the current one.
‘Last feature’ goes back to the feature which was offered before the current one.
‘Best’ makes current the best of the options in the choice model; usually, or when things are working well, the best yet. In the case that that option is already current, there may be a bit of update, otherwise nothing.
‘New option’ asks the id for a whole new option. What happens here is that the ego is attending to a feature when a whole new value for that feature is popped into mind by the id. The ego then says ‘yes, make a new option for the choice model which incorporates this new value for that feature’.
‘New feature’ asks the id for a whole new feature. What happens here is that the ego is attending to an option when a whole new feature for that option is popped into mind by the id. The ego then says ‘yes, make a new feature for the choice model which takes this new value for that option’.
‘Next option’ moves onto to the next option in the choice model, looping round to the beginning if the last option is that current.
‘Next feature’ moves onto to the next feature in the choice model, looping round to the beginning if the last feature is that current.
In this way, the ego jiggles about with the choice model until the current option seems enough better than others which it has looked at recently; a sort of local maximisation process. It then signals ‘yes’ and the id moves onto action.
In doing all this it will have regard to the urgency of the need for action.
This is the one place where the ego, where consciousness, has to been seen to be doing something, where it is not enough to be an observer on the unfolding scene; the ego has to respond to the id. It may be a fairly small function, it may not have much to do when compared with the id, but it does need to be there. An essential management function. I point to the analogy of a Prime Minister, touched on in previous posts: such a person does not need to do much, to originate much, but we do need one. See, for example, reference 2.
Summary
I have set out above the implementation of the choice function which I am claiming to be one of the two main points of consciousness. The hypothesis is that, somehow or another, the business of making these options for action the subject of experience is a necessary part of the ego making a choice of the sort outlined above.
We will be thinking about the other half of the suggestion at reference 1, the processes needed for the execution of decisions, in due course.
Reference 1: http://psmv3.blogspot.co.uk/2016/09/what-is-consciousness-for.html.
Reference 2: http://psmv2.blogspot.co.uk/2015/09/an-analogy.html.
Reference 3: https://www.bm-science.com/team/fingelkurts.html.
Reference 4: http://www.buzsakilab.com/.
Reference 5: http://psmv2.blogspot.co.uk/2015/08/the-freudians-fight-back.html.
Reference 6: http://psmv3.blogspot.co.uk/2016/09/the-choice-model.html.
No comments:
Post a Comment