ABSTRACT
The expression of attention to an object in the brain is assumed to involve increased activity at the location of that object relative to the activity in the surrounding area. The mechanism that is mainly responsible for this selective enhancement of activity is believed to be the thalamocortical circuit, which directly connects virtually every area of the cortex with a corresponding area within the thalamus. The operation of the thalamocortical circuit has been modeled by a neural network, whose operation shows that the circuit can simultaneously enhance and concentrate activity at a coded target location; in other words, the circuit has the capability of increasing activity at a target location while progressively reducing the amount of activity that is spread to surrounding locations. However, the equations of the neural network model involve a large number of parameters, and the solution of the equations requires numerical integration. We present here a model which contains many fewer parameters and whose equations can be solved analytically. This analytic model is based in part on the metaphor of a tubelight optical device that projects light energy upon a receiving surface, corresponding to the manner in which the thalamic neurons are presumed to project activity onto a cortical area. In particular, the analytic model mimics the neural network’s ability to simultaneously enhance and concentrate activity at a coded location. Predicted trajectories of target and distractor activity levels for both the neural network and the analytic model appear highly similar. It is concluded that the analytic model is potentially more suitable for empirical tests than the corresponding neural network model.
