poodleOwned
Posted : 1/29/2011 12:47:14 AM
Kevin Behan
This template can be shown to run through all behavior as in the dog/deer video linked on my site. I've also commented that given fractal mathematics and how it has been shown by Mandlebrot to be expressed everywhere in nature, I'm surprised that modern biology isn't concerned that no such fractal module in behavior has been identified.
Kevin, first up please understand that i am not trying to do anything other than to illuminate and speculate with this post. I am yet again going out on a limb. Mandlebrot showed how what seems to be a simple equation on paper can produce suprisingly complex shapes, and i guess that if we applied it to the right behaviours that it would tend to reinforce my points that much complex behaviour is a combination of very simple behaviours.
This simple equation is suprisingly difficult to understand. Most people have trouble with complex numbers. I used to spend several weeks trying to explain them to quite bright technicians. Much of say complex analysis looks (the basis to some extent of Mandebrot) very simple. It ties many experienced Maths people up in knots. Personally i like it. It is my kind of stuff. The problem with maths is that it says the same thing in many different ways. They are equvalent. An example is that we may choose to look at sine waves with magnitudes and angles or complex numbers . They are equivalent.
Well how does this relate to behaviour? One of the interesting things about Panksepp is that he has tied Affects to Anatomy. It is no longer speculative for a fair bit of what he says. Now taking my hat off as a trainer and becoming a cynic for a minute, my suspicion is that we may have missed some simple links for some behaviours becuase of data loss. Statistics is a kind of compression, a way of analysing things that are hard to analyse any other way. I think in this process that we could have lost three things, micro behaviours, time based information, and discarded sequences that might matter a dam. Now getting these behaviours requires intense possibly video analysis, right at the cutting edge o what we have now. The effects of Affects are very quick, and a sequence may produce behaviours that we would miss any other way.
I think that we have to acknowledge the neccessity of Behavourism, that it gave us some great pointers and ideas and was based on quite rigourous statistical analysis. My uneasiness has always been with the experiment design, and the underlying assumptions about body weight and the function of food to name a couple and of course the loss of data.
Skinner himself saw shortcomings. He said to Panksepp "A behavourial account has two unavoidable gaps - between stimulus and response and between reinforcement and a resulting change in behaviour. These gaps can only be filled with the instruements of Neurology"
P12 Panksepp "Affective Neuroscience"
I couldn't agree more, and i think this is where further analysis may be useful.
I think that you should also look at Wolframs book, A new kind of science.It is populist, he does over extend but he may make some points that you agree with. It is free to download. If you have questions about this stuff, i am always happy to help, either via a posting or via a PM. It is a genuine offer, as an Applied Scientist i have always spent a lot of time mentoring and educating . It is a kind of duty that comes with the knowledge.
Personally i try like hell to find some gem in what others say. I
work hard with much younger engineers trying to sort the great stuff
from the waffle, and have got quite good at it. I an struggling with
your NDT to a high level of frustration.