Within-level weights
All of the word units are mutually inhibitory (each word unit has negative connections to every other word unit).
In McClelland and Rumelhart's original IA model, there were also inhibitory connections within the pools of letter units in each spatial position. However, for the simulations reported by McClelland and Rumelhart (1986) the between-letter connections were set to zero, and so are not included in the BrainWave version.
