Nobody, including me, can know for sure what the choice is until I make it, and the choice depends on chaos. Even if it’s technically deterministic, it depends on how I resolve the noise that is emitted from chaos. If there’s true randomness in the world then that additionally helps me be the origin of the choice, rather than deterministic noise, but even with only noise from chaos rather than randomness, the rest of the universe cannot possibly know my choice until I stabilize on it because sensitive dependence on initial conditions means that the details that determine how my brain will wiggle around through neural consensus space are unobservable to any other system no matter how superintelligent, and the choice gets to depend on input from my entire brain. In this sense, my brain is still the causal bottleneck through which my choices depend, and my entire brain is the bottleneck; noise from chaos means that if I might have chosen a way that mismatches my full network of preferences, my neurons get a chance to discuss it before settling. Biases and shortcut reasoning bypass this partially, of course.
As a result, even if technically my choice is strictly a logical consequence of my brain state, that logical consequence is not written to the universe until I resolve it, and the chaos means that every physical system besides my brain must retain logical uncertainty about my choice until it is resolved which way my neurons discuss and settle. In a fully deterministic universe, free will is logical hyperstition.
Some interesting resources on the topic. I have watched the videos, but I only skimmed the search results. Bulleted summaries via kagi.com’s universal summarizer in ‘key moments’ mode.
The critical brain hypothesis suggests that the brain operates near a critical point, similar to a second order phase transition.
Second order phase transitions are characterized by a continuous change in properties, rather than a sudden jump.
The Icing model is a simple system that undergoes a second order phase transition and exhibits scale-free behavior.
Neuronal avalanches, or cascades of activity in networks of neurons, also exhibit scale-free behavior and are thought to be neural analogues to the Icing model.
The balance between excitation and inhibition is a key factor in determining whether a neural network operates in a subcritical, critical, or supercritical state.
The branch ratio, or the average number of neurons activated by a single upstream neuron, is a control parameter that governs the transition from decaying to amplifying activity in neural networks.
The critical point is where the balance between excitation and inhibition is optimal, allowing for efficient information processing in the brain.
If it’s new to you, I’d also suggest an overview of chaos theory:
Or if 10 minutes is a bit long, here’s a 1 minute animation showing divergence among chaotic trajectories that start out coherent; there’s a moment at :26 where the pendulums lose sync, briefly all at the same edge of stability; however, this is not a chaotic system which seeks the edge of stability, and the pendulums quickly fall in different directions off the saddle point. in contrast a system at the edge of chaos is on a saddle point at almost all times!
Nobody, including me, can know for sure what the choice is until I make it, and the choice depends on chaos. Even if it’s technically deterministic, it depends on how I resolve the noise that is emitted from chaos. If there’s true randomness in the world then that additionally helps me be the origin of the choice, rather than deterministic noise, but even with only noise from chaos rather than randomness, the rest of the universe cannot possibly know my choice until I stabilize on it because sensitive dependence on initial conditions means that the details that determine how my brain will wiggle around through neural consensus space are unobservable to any other system no matter how superintelligent, and the choice gets to depend on input from my entire brain. In this sense, my brain is still the causal bottleneck through which my choices depend, and my entire brain is the bottleneck; noise from chaos means that if I might have chosen a way that mismatches my full network of preferences, my neurons get a chance to discuss it before settling. Biases and shortcut reasoning bypass this partially, of course.
As a result, even if technically my choice is strictly a logical consequence of my brain state, that logical consequence is not written to the universe until I resolve it, and the chaos means that every physical system besides my brain must retain logical uncertainty about my choice until it is resolved which way my neurons discuss and settle. In a fully deterministic universe, free will is logical hyperstition.
Some interesting resources on the topic. I have watched the videos, but I only skimmed the search results. Bulleted summaries via kagi.com’s universal summarizer in ‘key moments’ mode.
The critical brain hypothesis suggests that the brain operates near a critical point, similar to a second order phase transition.
Second order phase transitions are characterized by a continuous change in properties, rather than a sudden jump.
The Icing model is a simple system that undergoes a second order phase transition and exhibits scale-free behavior.
Neuronal avalanches, or cascades of activity in networks of neurons, also exhibit scale-free behavior and are thought to be neural analogues to the Icing model.
The balance between excitation and inhibition is a key factor in determining whether a neural network operates in a subcritical, critical, or supercritical state.
The branch ratio, or the average number of neurons activated by a single upstream neuron, is a control parameter that governs the transition from decaying to amplifying activity in neural networks.
The critical point is where the balance between excitation and inhibition is optimal, allowing for efficient information processing in the brain.
Some links related to this summary on metaphor.systems—the ones I opened and skimmed:
https://en.wikipedia.org/wiki/Critical_brain_hypothesis—very short
Why Brain Criticality Is Clinically Relevant: A Scoping Review—interesting overview paper
Self-organized criticality as a fundamental property of neural systems—covered by the video thoroughly, but solid
Consciousness is a functional system that involves self-monitoring and hierarchical structure.
Consciousness is a complex dynamical system that emerges from the brain.
Embodiment plays a significant role in determining the kinds of conscious experience that we have.
Mental states are physical states.
There is an asymmetry between internal representations and what can be conveyed to others.
Chaotic dynamical systems are deterministic and unpredictable, making the question of free will unanswerable.
More metaphor.systems related results—the ones I opened and skimmed:
https://desirism.fandom.com/wiki/Brain-state_theories—hmm interesting site, I wonder if it’s any good
https://desirism.fandom.com/wiki/Free_will
The Hard Problem of Consciousness and the Free Energy Principle—not my favorite argument, the free energy principle is a coherent basis for doing further reasoning but is not in and of itself an argument for any particular view
https://en.wikipedia.org/wiki/Neuroscience_of_free_will—interesting, lots of stuff in here I didn’t know, I’ll have to go back over this at some point
https://plato.stanford.edu/entries/consciousness-neuroscience/ - looks very promising, but again, skimmed
If it’s new to you, I’d also suggest an overview of chaos theory:
Or if 10 minutes is a bit long, here’s a 1 minute animation showing divergence among chaotic trajectories that start out coherent; there’s a moment at :26 where the pendulums lose sync, briefly all at the same edge of stability; however, this is not a chaotic system which seeks the edge of stability, and the pendulums quickly fall in different directions off the saddle point. in contrast a system at the edge of chaos is on a saddle point at almost all times!