Compression and Concepts

ShareShare on FacebookShare on Google+Tweet about this on TwitterShare on LinkedInShare on RedditEmail this to someone

Earlier we discussed compression and stated that the intelligent mind leverages compression to represent information vectors more efficiently. Compression also comes into play in the development of concepts. Before I go into more detail on that I want to introduce another principle:

The intelligent mind leverages its senses to categorize similar information vector properties

I will explain this statement with some examples.

Suppose we put three colored disks in front of Billy: a red disk, a green disk and a blue disk. Although he has no concept of what we call “color” at this point, he does recognize differences in these three disks because his senses can perceive color (by the differences in wavelengths as described previously). The point above claims that Billy recognizes red, green, and blue as categorically similar things. Since different colors are perceived in a similar manner, the brain categorizes them as similar characteristics.

The same is true of sound: the brain will categorize two different volumes as categorically similar to each other and two different tones as categorically similar. This proposed principle is important because it supports the very important concept of comparison. And the comparison concept in most basic form is fundamental to the creation of other concepts.

Let’s say Billy was exposed to 20 similarly sized disks all of different color. He catalogs these items in his memory. What is the most efficient way for him to manage these memories? Since the objects are so similar, only varying in color, his mind will store one template definition that represents the disk size and shape, but he recognizes the twenty variations in the wavelength of the reflected light (what we call “color”). The recognition of the variation in this wavelength leads to the creation of the concept of color in Billy’s mind. We will see in general that the exposure to new information vectors will lead to the creation of concepts.

Now let’s say Billy’s attention turns to one of these disks, the green one. He notices that as time passes, the position of this item is incrementally different in that it is becoming further away from him. This changing of position is gradual in the sense that the new position is somewhat close to the previous one — the disk isn’t teleporting randomly through the dome. The individual memories he has of this observation over time could be described as:

  • Green disk is at one position at one time
  • Green disk is at incrementally different position at later time
  • Green disk is at another incrementally different position at another later time
  • Green disk is at yet another incrementally different position at yet another later time
  • And so on…

The constant concept in the bullets of is the green disk. The varying concepts are those of time and position. Compressing the observing into a concept:

The green disk changes position over time

Generalizing the change of position over time becomes a concept for Billy that we recognize as “movement”. So Billy’s memory holds the concept:

The green disk moves

Now let’s say that Billy observed the same behavior from the red disk, and the blue disk and the orange disk. In
this case, the movement concept is reused, but the each disk represent its own concept, so it’s the involved object that is the varying concept. However, Billy recognizes each colored disk as a “thing” in general. So his mind can compress these observations into a general concept:

Things move

The example above introduces the ideas of concept retrieval and application. This document discusses those ideas in more detail in subsequent sections. Once we explain concept retrieval and application we can discuss how more complicated concepts are created. For now, I will list some quick examples of fundamental concept creation:

  • The mind recognizes a similar set of tones that differ only in how loud they are. As other tones that also differ in this respect are experienced, the mind creates the concept of volume. Eventually when the mind experiences exact shapes of different size, the mind will create a generalized concept of magnitude that relates to size, volume, and so forth.
    The mind experiences many objects that have many of the same defining property, for example size, shape and color. By recognizing the measurable common characteristics across different things, the mind creates the concept of similarity.
  • The mind recognizes that individuals things are defined differently in many manners, whether it be the property values of the information vectors related to them or their position in space and so forth. The mind creates the concept of comparison to measure these differences.
  • By recognizing and observing one consistent thing and noticing that the value of any set of its defining properties differ across concepts of this thing, the mind creates the concept of change.
  • By recognizing a change that alternates between a set of values (for example, a blinking light differs in its intensity in a measurable on and off value), the mind recognizes the concept of pattern.
  • By recognizing that one concept directly and consistently leads to the triggering (for example, the pressing of a green button reliably initiates a musical tone), the mind recognizes the concept of cause and effect.

Ok, admittedly, the examples above gloss over a lot of the mechanical details on just how concepts are created in the processes described. Later sections of this document will go into more detail on concept mathematics. This document does not provide the blueprint for the algorithms that create concepts. It does attempt to introduce to a capable audience what is required to progress towards that end.

Previous: Billy’s First ConceptsNext: Retrieval of Concepts

Leave a Reply

Your email address will not be published.