Kauffman refers to the formation of ice as an example of a phase shift; however, the model could be more pervasive once you think of the possibility. Light switches are generally a phase shift event. That is, sequential pressure in one direction eventually produces a change in state, from on to off, that changes no further despite continuing pressure in the original direction. Phage shifts (gotta eat, gotta eat something different!) are another possibility as is the random event of a woman in love (she is or she ain't). Of course, there are intensities of love, an apparent contradiction to my suggestion of a phase shift. However, the ladies tell me of a distinction between "loving" and "being in love with." The former is analogue, the latter digital.
Phase shifts can perhaps be induced by the accumulation of output, call it waste sometimes. Output shuts down further reactions and creates a phase shift of no more activity until the waste scatters. It seems a variation of the negative feedback model or what Goodwin calls oscillation.
Why this phenomenon? Kauffman suggests, perhaps rightly, that any other model is inherently chaotic and will never replicate, never giving natural selection (at whatever level) a chance to operate. (1)
Not so incidentally, congratulations on your effort on behalf of Dr. Hebb. I'm envious, put me on the announcement list please.
NOTE:
1) My gosh, I'm writing this kind of stuff on a Wednesday morning? Time for risperidone! St John's would only make me worse.