The Adaptive-Habitual Intelligence
In any field of Science, we have the necessity and the luxury of organizing our study into a framework that makes it disciplined and documented. The basic dynamics that govern a phenomenon laid out, its structure or properties studied, its behaviors observed under different conditions, and theorized based upon the accuracy of test results, is necessary for the understanding of the subject. Probably one of the few exceptions, if not the only one, is Psychology.
Psychology is formally the study of mind and its behavior. Artificial Intelligence relies heavily on this meandering stream under Medical Sciences. The difficulty in studying a field like Psychology lies in the million-fold emergent behavioral diversities that arises in a population, or a sample space. The study of human intelligence, therefore, suffers much a similar state of existence. Let us concern ourselves with two broad areas of behavior here — adaptive and habitual.
Note: In Neuroscience, we could see how Cerebrum and Cerebellum in the brain, handle adaptive and habitual memories. We will discuss those details in one of the forthcoming articles.
Habitual behavior is our mechanism of carrying out a process that we are very well accustomed with. With possibly minor or least changes in environmental variables. Like driving a car or walking or typing out a letter on a typewriter. These are habitual actions, in our AI parlance, mechanized. We could go on doing the same and carry out with other activities that may require more mental investment, as if we were not doing anything at all.
Adaptive behavior is our faculty entrusted with taking care of situations and circumstances that are new, and hence require a more different approach. This is where every person could be more or less different, varying and the bearings of which on the person concerned depends upon the fabric of his mind — between a gradient of tough or meek, valiant or cowardly, firm or yielding, an infinitum of possibilities of combinations. Adaptive behavior makes the study of human mind as perplexing as it is mysterious.
The Adaptive-Habitual Connection
Adaptive behavior has its foothold on consciousness. The activation of multiple strains of nerves cells in our brains in a joint effort of waking up to the situation is an event that is special and vital. It is the most elaborate and ingenious mechanisms, and wherever it turns up, it rapidly acquires a dominating role. As we know it, it is a cerebral process that is associated with consciousness.
It needs to be known however that our habitual actions, those absent-minded activities we carry out, too are processes of the cranium and or nerves just as much. However, the effect of our conscious attentiveness is on a much-diminished scale. Logically enough, we need not tax ourselves as it is now known, and just needs the attention of probably fewer cells of the cranium and the lower nervous intelligence. It is only an act of resource sharing — the conscious mind is in command of the newer experiences, and as we get accustomed to the string of events, it slips slowly out of the domain of consciousness and into our mechanizable intellect.
The Mathematical Analogues of the Adaptive-Habitual
Habitual, as mentioned is mechanized behavior. If we are to synthesize a habit, we work out the optimal algorithm, or gradually polish our redundancies down to smoothness. We may model this behavior with the tools of Discrete Mathematics. We discern, we define and formulate the problem, solve, and implement the algorithm. At every step we have the element of predictability — discrete mathematics is largely deterministic, at every stage, we may know where the state of our mechanical system, just as we want it to be. It is possible.
Adaptiveness, on the other hand, is largely stochastic. A million-fold democracy of neurons taking part in one singular activity. Whatever the biological mechanisms involved; the activity of the neuron population contributes to the outcome. Like a species evolving slowly with time. As with any stochastic process, the process of consciousness and thought must fit into the framework of mathematical distributions and random processes. Say, like the seasonal migrations of geese.
The maximum order we could predict in the behavior of a distribution is with respect to the statistical tendencies that describe the population. Our subjective approaches to different stimuli are most likely the statistical behavior of the ensemble of our neurotic core. As a result, we shall have no clue as to what the state of the system is at a given time. The randomness of adaptive behavior would be taken care of.
The Intelligence System
The mathematics governing our adaptive behavior provokes us to push to stand by what John von Neumann says:
The defining characteristic of a complex system is that it constitutes its own simplest behavioral description.
Our brain is a very, very complex system. A system exhibiting adaptive behavior. And the simplest design of a brain is itself. The only practical way to simulate one such system is to build one. Also, to take into consideration the vast expanse of infinitum we have in the continuum of everything around us, we would need to think beyond the counters of discrete mathematics and the axioms of ordered logic.
Since current Digital Technology doesn’t consider the continuous nature of signals, vast possibilities of storing relational information and their processing based on context, we could assume that digital technology in the current form might not be suitable to replicate and simulate human brain and consciousness.
Our next best alternatives are, (1) pure mathematics to represent continuous and infinite information and (2) analog computing that can process this information without distorting the information by sampling. Hence our task explodes to qualifying consciousness in terms of differentials, integrals, exponentials, linearities and constancies, the possible variations in responses to environmental stimuli. We will also rely on Analog Computing to represent and process this information. It is complicated. But it is possible.
Before getting into the Analog computing and technology side of it, we need to dive into the complexities of human brain and understand how it works at a high level. Brain’s collective consciousness is the result of billions of neurons’ activity. Exploring biological side of the human brain will help us to apply those concepts into technology and get similar results — Machines with consciousness.
Next week: Neuron — The building block of Brain
Previous week: Mind and Matter — Consciousness
First week: Can Machines Think?
References :
