FIGURE 4.

A parallel forward models algorithm that generates levels of awareness within a system. This scheme is based on the concept that a system that can accurately predict its future state based on its current state has awareness that is functionally significant for the system. This form of awareness is postulated to be necessary but not sufficient for subjective experiences such as pain. This algorithm processes input simultaneously along three parallel streams. The base level is the processing module that creates an output for the system. A second level consists of a module that creates a simplified model of the base level processing that predicts (predictions 1) the future output of the system. That prediction is then compared to the actual output in a separate module (comparator) and the error generated is used to adjust the internal model 1 to better predict future states (by attempting to reduce error 1 to zero). A second internal model (internal model 2) uses both the input to the system and available inputs from other systems to predict the outputs of internal model 1. Prediction 2 is feedback into internal model 1 to facilitate a rapid and accurate prediction 1. Prediction 1 likewise feeds back on to base level processing to control its output.