The Discriminative Lexicon: A Unified Computational Model for the Lexicon and Lexical Processing in Comprehension and Production Grounded Not in (De)Composition but in Linear Discriminative Learning

Figure 13

The linear network corresponding to the transformation matrix . There are two input nodes, and (in blue),and two output nodes, and , in red. When the input vector is presented to the network, the value at is and that at is . The network maps the point onto the point .