Over the past year or so, I have been working tirelessly on the problem of how simple 2D pictures can be understood in the mind as 3D environments.

My work in general, is based on the belief that reality is adequately expressed by form and behavior, without the need for language or math abstractions. So the existence of a working engine is sufficient to define its identity, whether or not some equations exist that can describe it, or that bits can be labeled with language.

And that a real engine, made from matter, can be represented by information mimicking the form and behavior of that matter, and can thus model and predict it, free of the constraints from space and time found in the physical world. It is my belief that this is a precondition for consciousness. The problem has been, how do you cause 'particles' of information in the brain to express the properties of matter - such as adhering to neighboring atoms, momentum, inertia etc. To render an inner vision plane, and all this within a parallel processing environment.

After extensive efforts in the field of vision, I have been unable to discover any processing paths capable of this transform. So I have decided to go back a level, to where muscles interact with touch to discover the outer reality.

My theory is that the memory traces created by touch sensations over time, and the memory scripts that direct the muscle groups over time, can be considered as the foundation of a 3D 'vocabulary'. That reconstructions of the outer world of space-time can be simulated from those scripts - since a blind man can still understand reality without vision.

It is my proposition, that although vision provides the illusion of 3D space, it is actually acting more like a map, onto which the more fuzzy sensorimotor scripts can bind. The sensory and muscle scripts actually building the space/time dimension we experience. 2D vision simply overlays a map upon these existing sensorimotor traces, to give us that 'distant observer' perception of what we have already learned as feeling territory.

The vision plane acting both as a map for spatial precision, and as a method for organizing these sensorimotor scripts into a 3D scene, without the need for direct physical exploration.

If you see an apple, the recognition draws in sensorimotor scripts which draw a layer of tactile 3D 'object scripts' upon the vision. Any thoughts of interacting with the apple, being done in the sensorimotor domain, with the vision following. Since vision is very precise, the mind has a method of tricking us into believing we see more than we really do. The underlying touch and muscle action scripts are fuzzy enough to give confident prediction and interaction with reality withut the feeling of alienation. Physical movement also draws the concept of time into the simulation, since muscle and touch signals physically disclose the behavior of form over time.