Connecting units to each other

Hi there,

I’m working on developing a attractor network that requires the units of a layer to connect to each other, in addition to the output and input connections. I’m not quite sure how it can implemented without using a recurrent layer, or am I restricted to using a recurrent layer with some modifications? Thanks.

I’m not sure I understand the use case correctly. Could you post a drawing showing the schema you are working with?

Thank you for your prompt response.

The purpose of the network is to simulate how human acquire language (i.e., cognitive modelling). Schema as follows:

In particular, I’m having trouble trying to connect the Phonological Units to one another.