Friday, May 16, 2008

Why do we forget? One perspective

Another dose of musings of a wandering mind.

Why do we forget?
This, I think has more to do with gathering correct data about one's surroundings and skill honing. By nature related events or objects have the tendency of occurring close to each other have (spatial nearness) or are repetitive (temporal nearness). By remembering only events that occur repetitively or close to one another, one has a greater probability of having consistent and generally correct information about one's surroundings in one's mind. Giving this a more detailed look one realizes that repeated activity say like improving one's skills at a sport (practice makes perfect) are not possible without forgetting the wrong move and reinforcing the right moves with time. We with our set of neurons in essence keep an impression of repeated events/associations from the environment we are immersed in. Why is it important to remember events which occur close to each other in space or time? One reason: Things or events occurring close to each other, generally have a causal relationship. An organism having the ability to recognize such relationships has a tremendous survival advantage. By recognizing certain events that may trigger unfavourable events, one can avoid those, well in advance.

But, there are ways in which to give certain associations a higher weightage than this default mechanism. Emotions are a way to give certain experiences a higher weightage than the default weightage. for example, we don't learn that fire burns by repeatedly burning ourselves. One such accidental burn may leave a lasting impression (and a scar) to remind us to beware of such an event in the future. More over already made associations may help recognize the cause of such events. The sensation of pain gives the associated event a higher weightage for memorization. And, how did we initially start to feel pain - by mistake. There may have been freak organisms that felt no pain and walked fearlessly into fire. They most definitely got filtered by the sieve of natural selection. One mistake (mutation) by chance may have advantaged an organism of that species with the severe feeling of pain from burning. Such strong sensation evokes a strong emotional response making a lasting impression. Many such fears/feelings may have been incorporated in our DNA quite some time back, with the subsequent branches all sharing these. Fear, love, camaraderie are all emotions which may have altering effect on the weightage of the memories, and may bypass the normal repetitiveness and forgetfulness mechanism. Also, what remains after the years are those that advantage an organism in some way to survive.

This same mechanism built into us by nature - that is recognition of patterns and reinforcing of repetitive occurrences helps us recognize mathematical formulas, languages. There may have been fine tuning in the later stages of evolution. This pattern recognition and association of similar concepts is what makes some ideas appear so fundamental to us and many ideas not so tangible to grasp from our understanding of nature. Object oriented programming and many other programming concepts being cast as abstractions of real world objects comes from our understanding and comfort with analogues from nature.

This natural scheme of selective adaptation and a neural network capable of recognizing "important" patterns in the environment in which it is immersed in, is not only simple but very flexible. We don't expect fire to be cold. We grew in this environment. The hardwired feelings are only suitable to this environment. Feelings and emotions are a way to prioritize many of the experiences and translating them into increased weightages for the neurons they effect. Just immersing a blank neural network with forgetfulness function in a rich environment would retain associated events and objects because of their temporal or spatial nearness or higher frequency of occurrence. This observation of events and objects and their associations depend on the senses that pick these up and flood the network. An organism say with only auditory senses will be limited in its perception of its environment. But, if it survives, it may have gathered strange ways to prioritize what is important for survival. The interplay and associations between signals sent by various senses of an organism and the feelings they evoke - could be very interesting and could make for a colourful experience.

When thinking about intelligence. The first question that comes to mind is intelligent about what? Intelligent about one's environment? Intelligent in mathematics? Decoupling the environment of a machine or an organism and then talking about intelligence does not quite make sense. Or does it? After a certain amount of complexity is attained, an organism may in a sense "live in its own mind" - detached from nature, having the ability to see ideas and associations between them with no real analogues in nature. This may reach a point where now the being may want these ideal structures held in mind to be translated to reality or checked against reality. One is free to have anything in one's mind. But, an organism detached from the environment cannot survive long. If the organism is "unlucky" sooner or later the organisms ideas are tested against reality. That's where Science comes in where ideas get validated and tested against reality, from time to time. Some theories remain and some fail the test. A race has a better chance of survival if the knowledge it acquires is correct and consistent with what is real.

No comments: