In a talk for the 2024 SFI undergraduate research program, Simon DeDeo described how instead of interacting directly with reality, the mind interacts with “the symbolic,” a third thing used for seeing through. This is necessary because the real—in the Lacanian sense of what is resistant to formalization—is too big for our minds to process.
(Tangent #1. Compare this with Life is a lineage of information, after Walker: because life is contingent on structures that have already been built, there exists an “adjacent impossible” of things that we cannot conceive of. However, Walker also claims that imagination causes the reality that will be built, and the universe we imagine is bigger than the universe we can physically construct. I have yet to square these ideas—is the signature of life to reshape reality such that it squares precisely with our mental models?)
(Tangent #2. More generally, internal models are used by all complex adaptive systems: “Everything adaptive has a model.” How does the sense of reality being “too big” differ across adaptive systems, if at all? …What is the relationship between a mental model and unconscious thought?)
There are two “oppositions” in how we use mental models:
How do we get a grip on this “third object”? Take mathematics as an example. While trying to prove a theorem, a mathematician interacts with the space of all possible proofs by building a mental representation of the space of proof structures. If mathematicians are working collaboratively, like in the Polymath project, they must also model each others’ possible models. Modes of communication—speech, text, gesture—not only enable sharing models, but also talking about and assessing models (see also: Self-reference). Mathematical subfields exemplify models shared between individuals (though to what extent?).
But what about models that are not shared wholes, but distributed? Wolpert & Kinney (2024) essentially theorize that mathematical ground truth is such a distributed model (then use a mathematical model to get a better grasp on the implications of that theory…like a Russian doll!). This truth is clearly distributed because there is not one mathematician that knows all of mathematics, but is it actually a model? Perhaps this is a case of the mental representation squaring precisely with “reality,” to the extent that math can be considered real (alternatively, calling this a model is a category mistake).