Table 5.

The different training regimes used in VisNet experiments 1–4 of Section 5.4.5.

	Layers 1, 2	Layers 3, 4
Experiment 1	Trained on pairs	Trained on triples
Experiment 2	No training	No training
Experiment 3	No training	Trained on triples
Experiment 4	Trained on triples	Trained on triples

In the no training condition the synaptic weights were left in their initial untrained random values.