I have for years been haunted by an idea I call PatternResonance. Essentially, one begins with the ideas in signal processing, and in particular the issue of resonance. In signal processing theory, an input signal is transformed into an output signal. Of fundamental importance in discrete time and digital signal processing is the linear time invariant transform. This has a well defined notion of a frequency response (since the eigenfunctoins of a linear time invariant transform are complex exponentials, whose real parts are thus naturally sinusoids). This is how things are when we have a single input and a single output. Arrangements of a small number of signals are commonplace in modern music production. What interested me was applying these ideas to structures such as the human brain. A natural way is to consider the output of a neuron as a signal, and thus to have a signal processing system with a massive number of interrelated signals, and transforms applied to multiple signals rather than just one. Something like that. Then we can start to consider how an impulse, or

*pattern*of impulses decays through the resulting network. For now I shall dump some of those ideas here.