Hi John Hi Carmen, In your post, you state... "The number of explicitly chunked forms of perception is irrelevant to the endeavor of modeling - this is the point of using transferability as an evaluation metric. Along with the succeeding phrase, this also seems to suggest conscious process - which we expressly argued to be inappropriate until the point where the modeler can reproduce the same consequences in the world as the original model." Exactly. The number of explicitly chunked forms is always irrelevant. However, the *ability* to chunk forms is both relevant, moreover imperative, in the conscious coding stage of the modelling process. Otherwise, with no ability to code - transferability as an evaluation metric is inappropriate and possibly useless. Like the language of our species, transferability of perception, meaning and intention depend on the ability to code (unconsciously initially / consciously secondarily) those perceptions, meaning and intention. However, I would agree on your point that it is possibly more useful than not to suspend conscious filters in the unconscious uptake stage of modelling (ie-Nerk-nerk state) prior to 'attempting to codify it.' I am curious about... "... but it seems quite clear that anyone with a third also has reflexivity **while it is perfectly possible to have a 2nd without reflexivity.**" (emphasis added) What would be an example of having 2nd position without reflexivity? Thank you for your continued patience and posting, be well, Gene Bryson