This never occurred to me before, but the self-acknowledgement that one has successfully enacted a meme must often come from within, i.e., there must be some “self-triggering” of the memetic reward facility in there somewhere. This may or may not (need empirical data) be closely tied to the ability to recognize and reward when another successfully executes the same meme. It may, it may not, it may be a bit of both, it may vary a lot from person to person — experiments are needed.
This self-rewarding for enactment of a meme, or even just the recognition of a meme in the environment may be basic to taking a memetic inventory of a given cohort. The fact that one can recognize when one has enacted a meme correctly oneself is part of recognizing whether somebody else has done it. All of this is incidental to how much or how correctly a given meme is actually being enacted in the environment of the cohort.
Why Ideomemetic Self-Reward Memes?
One has to be able to give oneself feedback as to whether one is enacting a meme properly. In fact, these clues may be more “detailed” than what is generally perceived as comprising the meme. There may be more miniature intermediate rewards built in to an ideomemetic representation of a meme, moreover.
What if there were no ideomemes? One would get no pleasure from doing something when nobody else was around (with a few obvious exceptions, of course). Learning an activity by oneself, practicing, etc., would all look differently. There would be enormously more residual memetic debt.
 Two very different things?