I am now working through Reza Negarestani's book Intelligence & Spirit which deals with a lot of concepts, but one i think may be useful here is what he refers to as "Toy Models" and, within them, "Meta-Theories".
So a part of his project is creating a theory of the development of intelligence (here meaning strictly the interaction between subjects and the intelligible world), and a thing he thinks is important is asking the right questions regarding the development of AI. To that, he puts i that human intelligence should be the model of AI, on the level that we understand that our view of ourselves should be open to change and redefinition. He puts it by saying that Humanity is like a "toy model" (a concept he gets from physics i believe) for AI, and he puts it like that because "toy models" make explicit the meta-theories that we take as implicit in everyday life. In simplifying the theory of living and being a being imbued with intelligence in these models, we can take a step back and see the kind of pre-conditions, structures and transcendental meta theories (here being economy, capitalism, gender, what have you) so that we can properly separate what is natural and what is arbitrary.
So how does that relate to memes? To me, i think that meme-making comes with, if not this sort of semantic distance, at the very least, a new set of rules to normal communication that make it so our standard living becomes a bit warped.
With memes, we always assumed that some level of unoriginality is in effect (we're always re-telling or reworking the same jokes over and over again). At the same time, nothing is more glaring as a social faux pas than posting a dead meme. So the way in which we think we act as consumers is shown in its hyperaccelerated form (we are always sold things with slight, incremental changes as something new and exciting - cf François J. Bonnet when he talks about the idea of synchronized living in the digital age). Though most meme-makers and meme-sharers end up reproducing this cycle, i feel like there is a personal level of whiplash in realizing that a meme you are still taking enjoyment out of is dead. I personally noticed how many meme pages ended up making fun of the "they're doing surgery on a grape" meme were sharing them like crazy before - there is a degree of self-hypnosis into being "with it" that i think is interesting to look into.
I feel like Memes being a shared language is a level in which they exist as collective but not quite as a community, which i think makes projects regarding to them a bit hard. I think by design, they have two very important formal structures :
1 - They create an in-group and out-group mentality - there is the ones that get it and the ones that dont (i think looking into the facebook group What Does It Mean? is very useful - it is a group where people share posts, jokes and such that they dont understand in the hopes that someone explains them. Seeing these formal aspects broken down into sort of "plain-text" create that effect that, well, explaining a joke makes it unfunny)
2 - They're nearly by definition very dense - meaning that they compress as much referential power in as little space as possible. Which in turn means that if there is a missing link in the chain, it all kind of falls apart. I think this is part of why people make fun of leftists memes being walls of text - they break that rule of compression!
Those two structural aspects i believe make it so communities can get stronger with the spread of memes, but the spread itself ends up causing a qualitative differentiation between who gets it and who doesnt. Its probably not impossible to build communities with them, but i think their strenght lies in reinforcement.
In any case, very good luck on your findings!