Some developmental and attentional biases in the contrast enhancement and short-term memory of recurrent neural networks

Author(s): Grossberg, S. | Levine, D. |

Year: 1975

Citation: Journal of Theoretical Biology, 53, 341-380

Abstract: This paper studies the global dynamics of neurons, or neuron populations,in a recurrent on-center off-surround anatomy undergoing nonlinearshunting interactions. In such an anatomy, a given population excitesitself and inhibits other populations. The interactions are defined bymultiplicative mass action laws. Grossberg (1973) studied the case in whichall populations have the same weight (or total number of unit cell sites).Here the effect of an arbitrary distribution of population weights is studied;each set of populations with equal weight is called a subfield. Possiblecauses of variable population weights are developmental biases (e.g.,which feature; detectors are represented in a field), attentional changes(e.g., which features are relevant at any time), and statistical errors innetwork design. Such factors can bias the total field towards accentuatingor suppressing in short-term memory a given subfield of sensory features.In particular, a mechanism is noted for suppressing the activity of populations whose trigger features are infrequently experienced by the network.These variables interact with the recurrent on-center off-surround interactions, that have previously been shown capable of contrast enhancingsignificant input information, sustaining this information in short-termmemory, adapting the field's total activity while producing multistableequilibrium points of this activity, suppressing noise, and preventingsaturation of population response even to input patterns whose intensitiesare high.

Topics: Mathematical Foundations of Neural Networks, Models: Other,

PDF download




Cross References