Information Vectors can be composite — one information vector can be made up of other component information vectors. An eight ball is a black sphere with a small white circle on its surface and a black number eight painted within that white circle. The three separate information vectors: the black sphere, the white circle on the surface and the number eight are components of the composite information vector: the image of an eight ball.
The idea of compression previously discussed contributes to vector composition. If somebody briefly displayed something that appeared as an eight ball to you, you would identify it as an eight ball. Upon closer examination, you find that what you perceived as the number eight was really a “nearly closed loop” three. Your mistake was an understandable one since the object was black, and the size and shape of a billiard ball, and you had never seen a black three ball before. Seeing all these vectors together led you to believe, erroneously in this case, that you were seeing an eight ball. This example touches on the ideas of information vector comparison and concepts which have not been detailed yet. The important point for now is that compression supports the efficiency of information vector composition.
Here is a variation on information vector composition. How would you describe this image below?:
It is likely that you would describe it as: the letter “H” with an erasure line drawn through it, or something to that effect. You wouldn’t perceive and mentally model it as the four detached black segments aligned in a certain order. So this information vector is a composite of an “H” and a continuous white erasure line that runs through it. Even if you don’t know what the letter “H” is, I will argue that your mind would still model this information vector as something “H-shaped” interrupted by the line because you prefer to perceive the continuous shape. Incidentally, this thought is in line with the Gestalt law of continuity.
When the perceptive mind interprets composite information vectors, it has to decide which information vectors are composed of which other information vectors. Take this example: if an observer with no concept of trees and squirrels sees a maple tree with a very still squirrel climbing on it, he may take the “squirrel vector” as an attribute of the overall information vector. It is only continued perception of trees and squirrels on their own that leads the knowledgeable person to understand their individualism. But this point also deals with the topics of concept and information vector comparison, both of which have yet to be discussed.
One final thought for this section: composite information vectors should really just be considered information vectors. A good majority of the inputs we perceive are composite information vectors. The point of this section is that information vectors can be comprised of other information vectors.