It was the process of evolution of mankind which has determined what we can consider as the current design plan of the human species. Undoubtedly, the evolution has generated an outstanding excellence in cognitive performance, thereby adapting to the environment encountered. This is unsurpassed so far by anything we can think of. Nevertheless, there is a certain sub-optimality concerning our performance in modern times. The slow pace of natural evolution lets us assume that the human race of today mirrors the predominating needs of survival at times about hundreds of thousand years ago, but not those of today. This evolutionary adaptation process has not been able to catch up to all the demands humans are facing in modern times with rapidly changing work environments corresponding to the steadily advancing technical equipment. As a simple example, deliberately not considering latest technical developments, the vehicles humans are operating generate speed levels which are by far no more comparable with the running speed the human can accomplish without any supporting technical means. An equivalent decrease of human reaction time necessary to master sudden encounters of threats has not evolved simultaneously. Thus, there is a price to pay for the fact that the environment of human operators has changed, mostly by humans themselves. Therefore, the work system designer has to account for both the excellences of human cognition and the weaknesses as they exist here and now. Thus, let us move to more details about the features which are of interest for the work system designer. We know that the topic of human cognition is at least worth a study for itself, like those which are available. Nevertheless, it makes sense to introduce this topic here by selectively highlighting what is of fundamental interest for the designer of artificial cognitive components in work systems. That is the discussion on:
• connectionistic information processing,
• division of cognitive functions in distinctive network structures,
• the two modes of information processing,
• principles of human memory, and
• the ventral loop.
The main brain structures are described in some depth from the anatomical and functional point of view for those who want some more biopsychological background for what will be outlined in the following. Since biological systems are principally based on chemo-physical processes, the solution for the implementation of cognition in the human brain presumably cannot be but a connectionistic one. The functional elementary building blocks for cognition are nerve cells (neurons) in a highest possible packing density. The neocortex with its two hemispheres, for instance, the largest of the main subdivisions of the brain (90% of the cerebral cortex, also called isocortex because of its more or less continuously layered networking structure of neurons) comprises about 1010 neurons. In anatomical terms these are packed in a flat sheet with a thickness of about 3 millimeters. It is considerably convoluted in order to fit into the skull, including all neural interconnections. The main furrows (fissures) of this folded sheet spreading out over an area of about 0.2 squaremeters give a good indication about cortex divisions, resulting in four main lobes of each of the two hemispheres: frontal lobe, parietal lobe, temporal lobe and occipital lobe. The neurons are interconnected to a very high degree (about ten thousand interconnections per neuron) and mainly organised in columnar arrays. There is the principal property that these structures are not invariant in the course of the human individual's life. The brain adapts to changing demands by increasing or decreasing the number of neurons and/or interconnections to a certain degree. The interconnections take care of the transmission of the output signals of neurons to pertinent destination neurons which in turn are processing the incoming signals and are producing their own output which may be widely spread out. It is noteworthy at this point that despite the enormous performance we demand from this neural supercomputer the energy consumption is extremely low, only at about 15 Watts.
There are certain mechanisms located at the connection points for incoming signals into the nerve cell, the so-called synapses, which determine the contribution of these signals to the output of the neuron. The synaptical parameter setting represents the status of embodied knowledge available for a human individual. This knowledge is the main determinant of human behaviour. The synaptical setting is modifiable through the process of learning. The connectionisitic implementation lends itself to integrate the ability to learn. Our cognitive performance would be rather miserable, if we could not rely on our learning capabilities. This is fundamentally different to artificial cognition. Knowledge cannot easily be "implanted" into the human brain, at least not at the time being. Therefore, we are permanently learning in our life. We never arrive at a stationary state of accumulated knowledge. We can push the learning process by conscious training, which usually is a very tedious process, otherwise we learn by in-built automatic learning mechanisms which rely on a great number of repetitons of presentations in the course of normal operations. Learning not only relies on changes of parameter setting of the synapses by these mechanisms, but also on a certain amount of plasticity of the neural structures by adding and closing down interconnections. In addition, changing demands might also lead to increase or decrease of the number of neurons. This accounts for the so-called plasticity versus stability dilemma by increased limitation of the plasticity with the individual's age.
Our website is not responsible for the information contained by this article. Articleinput.com is a free articles resource thus practically any visitor can submit an article. However if you notice any copyrighted material, please contact us and we will remove the article(s) in discussion right away.
Note: This article was sent to us by: Samantha E. Dodson at 01122010
1. Regulation of synthesis and secretion
All articles are property of their respective authors. Please read our Privacy Policy!
© 2009 ArticleInput.com.