Wednesday, 18 July 2012

Feedback in complex, dynamic systems

One essential feature of complex, dynamic systems is feedback and the mind is no exception, including the higher scale of society.

These systems contain functional elements that produce output to their environment, the environment is modified to some extent, and the resultant conditions impact on the system in turn, which then produces some output once again that feeds into the surrounds, and so on and so on.

To what extent does or can the feedback alter the system?

Since a complex system is complex no matter how intelligent its members are (even an ant hill is a complex system) the eventual result depends on the members to process the information. There may be exceptional members, but, if they are part of the feedback, the question revolves around the degree to which they are understood by the rest.

Their contribution may not be understood at all or it could be misunderstood. In either case it may not be acted upon out of fear of the unknown. Assuming a certain developmental curve towards increasing complexity, there is a time lag between what is not taken up and what is taking part in the feedback loop in any case.

If the latter is sufficient to sustain the exceptional contributors, the system will benefit from them because when the rest finally do catch up those contributors are still around. A similar effect is gained through the welfare system in those countries which can afford it. The exceptional contributors don’t survive because society values their input, they survive because they are kept alive anyway.

At the same time however the general wealth of the society making this possible needs to come from somewhere and in a competitive world a nation needs a sufficient degree of intelligence to stay ahead – which means it needs to be able to productively absorb the feedback and the more esoteric contributions (for want of a better word) may not be necessary to maintain the relative advantage.

And here is the rub: anything that gets assimilated and thus is enhancing the system’s validity has therefore a considerable impact; but likewise, a negative addition will have a destructive influence in the longer term.

Since feedback is self-defining, once a downward path has been embarked upon it becomes harder and harder to compensate for – the feedback itself will have become diminished.
A precarious phase has been entered from which it becomes impossible to recover.

Civilisations that have fallen followed that pattern. One of the most detailed descriptions of such a scenario is Edward Gibbon’s “Decline and Fall of the Roman Empire”.

The gradual decrease of the complexity inherent in the information that gets passed around creates constantly diminishing responses which in turn generate poorer material for the feedback loop. Not only has the overall standard deteriorated, a relatively higher contribution has a greater chance of meeting opposition and the degree of the latter’s intensity becomes a measure of the conceptual distance between the exception and the norm. The result is a accelerating deterioration of the overall mindset.

A reversal is unlikely unless the prevalent feedback mechanism is unseated through the interference by a sufficiently powerful authority. Since this type of interference requires a structural framework to perform in the first place, ordinary dynamics in terms of information flow will have been affected too – the system has changed in any case.

Examples from the intellectual realm would be inventions that were dismissed because the general ambience had no room for them. They can also be found in politics, although here the sections within a party must be considered as they could play the part of the higher authority. A similar role can be played by polls which can create an affinity with certain sections of a party who could feel empowered by them.

Examples can also be found in the biological sphere in general where the complex dynamic framework is represented by an eco-system and its members are the flora and fauna. In this case the information flow does not consist of cognitive data but of the resources such as food which must be attained and needs dispersal in order to proliferate, and which can improve or deteriorate under certain conditions. Abrupt changes can come from climate (eg, floods or drought), a sudden change in the balance of plants or animals (eg, through disease), or the availability of food in general which may support one life form over another, and that in turn influences the eco-system from then on.

Furthermore, steadily worsening conditions are harder to turn around because once the previous equilibrium has been demolished a certain re-adjustment process has set in which actually mitigates against a reversal. The system as a whole has become inherently unstable; at that point anything can happen.

The question is, can such a state be regarded as a bifurcation under the terms of chaos? After all, it is not a change from one particular pattern to another; there is no other pattern.
Is this therefore an ‘extended’ bifurcation, a dynamic space which is pattern-less, has no discernible hierarchical order of any kind, and therefore has no ‘type’?

The next question is, for how long can such a state persist?

The usual determinants of impact, scope and size of its constituent elements are no longer applicable, nor can anything be gleaned from observing the in- or output. For that kind of analysis to be useful there has to be a pattern, yet this is exactly what is missing.

Perhaps this is one situation that most closely resembles what is commonly understood by ‘chaos’, a featureless, dynamic state. Just like the brain at its inception.

This also offers an answer to the previous question. The reason the brain can change from its featureless state to a more organised system is through constantly repeated input such as parents provide for their child. Therefore the one path out of that dilemma is constancy. Constancy in terms of input, overall conditions, and the availability of resources the system needs to sustain itself. Remove any one of them and the system cannot re-form, or form altogether in the case of the infant’s brain.

In principle the nature of that constancy does not matter except in the sense of how the system is expected to perform eventually. For example, abandoned children cannot fully attain their ‘humanness’ if left outside human society for too long, where ‘too long’ refers to the time spent laying down the alien patterns. After a time the already formed internal structures cannot be undone; a phenomenon characteristic of progression lock (the constraint imposed on new developments by already established patterns).

Considerations about feedback loops, continuing to bifurcations, and then on to instability and progression locks are part of complex dynamic systems such as the mind.
The fact that they can be entertained at all demonstrates the interdependency within these systems.

No comments: