PredictCorrectSetUp sets up the simulated space to simulate the responses of PBN neurons using the target-tracking predictor-corrector model. A predictor-corrector model system without an observation could not generate a correction, but it could fall back on its prediction to estimate target position. According to the assumptions of the model(model of target-tracking by animals with an SSC activity independent of the target like if the lights were turned off while a cat was tracking a light), if a predictor-corrector system for target tracking exists in the brain, then it should continue to produce target position estimates even if visual(or other sensory)observation of the target is discontinued. In the example of the cat, the parabigeminal nucleus(PBN), receives input predominantly from SC. The firing rates of PBN neurons continue to increase as retinal position error increases during brief periods when the moving target is not visible. The firing rates of single PBN neurons are not linearly related to retinal position error. Instead, responses of PBN neurons either fall off or plateau(saturate) as the target moves and retinal position error exceeds a certain value. Thus, their responses may be proportional not to target position itself, but to the probability that target position falls within a range, or exceeds a certain position that is specific for each PB neuron.
Anastasio, Thomas J. Tutorial on Neural Systems Modeling. Sunderland: Sinauer Associates, 2010. Print.
Researchers should cite this work as follows: