ABSTRACT

This work builds upon the proposal of the Boundary Contour System proposed in [GrossbergMingolla85] and [GrossbergMingolla87] as a subsystem of the FACADE architecture introduced in [Grossberg90]. We analyse the positional sharpening capabilities of the competitive layers in the competitive cooperative loop. Positional sharpening as set up by Grossberg and Mingolla may be seen as a two-stage process, where a threshold-linear signal function transforms the input and drives the subsequent positional competition stage. We draw the conclusion that positional sharpening is enforced by thresholding rather than by positional competition.