Progressive attractor selection in latent attractor networks

Latent attractor networks are recurrent neural networks with weak embedded attractors. The attractors bias the network´s response to external inputs without becoming fully manifested themselves. Latent attractor networks have been used to model context-dependent spatial representations in the hippocampus, and to encode context-dependent stimuli in neural networks. In the current model, the selection of the biasing attractor occurred in response to an initial triggering stimulus indicating the context signal. For example, the sign on a door may set the context for the representation of a room. However, in many realistic situations, context is set by a set of cues rather than a single cue, and these cues are typically seen sequentially, though not in a particular order. The problem addressed here is: how can a latent attractor network progressively select an attractor in response to a sequence of context patterns?