I show here that two interpretations of neural maps are closely related. The first, due to Kohonen, sees these maps as forming by an adaptive process in response to stimuli. The second—the minimal wiring or dimension-reduction perspective—interprets the maps as the solution of a minimization problem, where the goal is to keep the “wiring” between neurons with similar receptive fields as short as possible. Recent work by Luttrell provides a bridging concept, by showing that Kohonen's algorithm can be regarded as an approximation to gradient descent on a certain functional. I show how this functional can be generalized in a way that allows it to be interpreted as a measure of wirelength.

I describe a local synaptic learning rule that can be used to remove the effects of certain types of systematic temporal variation in the inputs to a unit. According to this rule, changes in synaptic weight result from a conjunction of short-term temporal changes in the inputs and the output. Formally, This is like the differential rule proposed by Klopf (1986) and Kosko (1986), except for a change of sign, which gives it an anti-Hebbian character. By itself this rule is insufficient. A weight conservation condition is needed to prevent the weights from collapsing to zero, and some further constraint—implemented here by a biasing term—to select particular sets of weights from the subspace of those which give minimal variation. As an example, I show that this rule will generate center-surround receptive fields that remove temporally varying linear gradients from the inputs.