a fascinating aspect of biological systems is how well they can be modeled using cellular automata; after all, multicellular organisms consist of (nearly) identical units reacting to adjacency-dependent environmental conditions in ways which generate emergent properties
---
RT @ulkar_aghayeva
"How to understand cells, tissues and organisms as agents with agendas"
h/t @RogersBacon1
https://aeon.co/essays/how-to-understand-cells-ti…
https://twitter.com/ulkar_aghayeva/status/1444812107040575488
the bit about blocking intercellular gap junctions causing a breakdown in collaboration and triggering cancerous growth reminds me of certain aspects of autistic behavior, namely how agents unable to communicate effectively end up with a radically different conception of selfhood
more generally, there seems to be a societal parallel; the rise of atomization has harmed our interpersonal communication, lowering social synchronization rates and driving individualistic modes of being
could improving communication skills, similarly, reverse this trend?
gap junctions are comparable to interpersonal relationships; both are communication channels between agents, and as such can be quantified in terms of bandwidth, with the two terminals of the spectrum being atomization and unification
how much bandwidth is needed to collaborate?
too little, and the agents cannot adequately synchronize on goals, and hence, align
too much and the agents cease to be distinct entities, which is likely suboptimal, as it lowers redundancy and parallelism
presumably related to the complexity of the relevant domain
the larger & more complex an agent is presumably the more bandwidth it requires to adequately align w/ another
I'd imagine this is proportional to the surface area of the agents informational boundary, as internal complexity increases w/ information flux
what's the growth rate?
not really going anywhere in particular w/ this line of thought, but I do think it's worth pointing out how easily the computational frame lends itself to application in the biological domain
information dynamics rule everything around me & cs/info theory are the study thereof
seems like synchronization is necessary to share internal state, intermediate results, threat intel, etc. this is a distributed computation problem; cells are not partition tolerant!
I wonder if there has been work done to model cell network info flow as distributed systems 🤔