Awakening From the Meaning Crisis by John Vervaeke, Ep. 31 — Embodied-Embedded RR as Dynamical-Developmental GI (Summary & Notes)

“Sometimes what makes something relevant is how it’s the same, how it’s invariant. Sometimes what makes something relevant is how it’s different, how it changes. And you have to constantly shift the balance between those because that’s what reality is doing.”

(In case you missed it: Summary & Notes for Ep. 30: https://markmulvey.medium.com/awakening-from-the-meaning-crisis-by-john-vervaeke-ep-b7155e7a09bd)

Ep. 31 — Awakening from the Meaning Crisis — Embodied-Embedded RR as Dynamical-Developmental GI [57:28] https://youtu.be/gfKcVbNd7Xc

• “There is a deep dependency between your cognitive agency as an intelligent problem-solver, and the fact that your brain exists within a bio-economy.”
• “The body is an autopoetic bio-economy that makes your metacognition possible.”
• No body, no mechanism for the process of relevance realization
• The biological fittedness of a creature is not a property of the creature per se. It’s a real relation between the creature and its environment. Fittedness is not a property of objectivity or subjectivity, it’s a property that is co-created in a dynamic, evolving fashion
• Vervaeke argues that we should not see relevance as something we subjectively project, as the Romantic claims. Instead, that relevance is transjective. (Neither projected or detected, but realized)
• Realization has two aspects to it: an objective sense (makes it real), and a subjective sense (coming into awareness).These two things represent the transjectivity of relevance realization. So it is necessarily both embodied and embedded.
• “When I say internal, I don’t mean subjective. I don’t mean inside the introspective space of the mind. I mean internal to an embodied, embedded, brain-body system. An autopoetic system of adaptivity.”
• The work of Marcus Brede shows us that when creating network — especially neural networks — you optimize between efficiency and resiliency.
• Internal bioeconomic properties vs. external interactional properties. The hand is a good general purpose machine, “but the problem with being a jack of all trades is that you’re a master of none.” So for specific tasks, the hand can be outcompeted by a special purpose machine. But you don’t just want a motley collection of special purpose machines. “Sometimes you’re adapted by being a general purpose machine, and sometimes you’re adapted by being a general purpose machine.” You want to be able to move between these two things.
• Special purpose machines are not used all the time, and the problem with that is very inefficient. But this makes you very resilient, because you’re constantly confronted with challenges for which require different tools to solve.
• How would you want to make information processing more efficient? You would want the functions — the processes — you’re using to be as generalizable as possible. (For a clue, look to A.I. researchers. They’re already making use of some strategies!) One thing we’ve learned about in statistics is the “line of best fit.” Drawing a line through a scatterplot allows us to interpolate & extrapolate — to make predictions. To go beyond the data. (“All good science takes good chances” was a great insight of Karl Popper) You can start to generalize. This is data compression, where you try to pick up on what is invariant and extend that.
• What about the opposite? Particularization. Where you create a function that over-fits in some sets and tries to keep with the data and stay more specifically in contact with the particular situation.
• Compression tends to pick up on what’s invariant, but particularization tends to pick up on more variations. Contextable vs. context-sensitive, dynamically trading between one another, between efficiency and resiliency. a.k.a. scope vs. applicability. When these two things are coupled they can help fit you well to the environment.
• These two things (scope & applicability) are cost functions.
• Explore vs. exploit. The more you stay put the more opportunity cost you accrue. But the more you move around the less you can draw from the environment. So you’re always trading between exploring and exploiting, and you can reward this either error reduction or error increase to keep this process going.
• This is known as cognitive tempering. It has to do with the projectability of your cognitive processing.
• These examples aren’t exhaustive but they are exemplary of virtual systems that can adapt within constraints between the sensory-motor loop and the environment. Sometimes you’re focusing, and sometimes you’re diversifying.
• “Sometimes what makes something relevant is how it’s the same, how it’s invariant. Sometimes what makes something relevant is how it’s different, how it changes. And you have to constantly shift the balance between those because that’s what reality is doing.”
• When you’re doing data compression (i.e. focusing on what is invariant) you’re focusing on how you can integrate, and when you’re doing data particularization (i.e. focusing on what is variant) you’re focusing on how you can differentiate. Piaget called the former assimilation and the latter accommodation.
• “If we just pursue efficiency — if we just assimilate — our machinery gets brittle and distortive. It has to go through accommodation, it has to produce variation, it has to rewire and restructure itself so that it can again respond to a more complex environment.”
• “When a system is self-organizing, there is no deep distinction between its function and its development. When a system is simultaneously integrating and differentiating it is complexifying. As systems complexify, they self-transcend. They go through qualitative development”
• As self-transcendent systems complexify it leads to emergent abilities. (“When I was a zygote I could not vote. I could not give this lecture.”)
• The findings of Spearman from way back in the 20’s was that how kids were doing in math was actually predictive of how they were doing in English and even how they were doing in sports etc., and vice versa on all counts. There’s a huge inter-predictability between how you do in all these different tasks. That is your general intelligence. (GI) This is the capacity that underwrites you being a general problem-solver. (Note: this is one of the most robust findings in psychology. It’s a powerful measure and it keeps getting replicated.)

• Vervaeke argues that relevance realization is the underlying ability of your general intelligence. “That’s how we test for it.”

Next up: Awakening From the Meaning Crisis by John Vervaeke, Ep. 32 — RR in the Brain, Insight, and Consciousness (Summary & Notes) https://markmulvey.medium.com/awakening-from-the-meaning-crisis-by-john-vervaeke-ep-32-e38bcd7214f8