Free Will - A Testable Hypothesis
Free will is not about control. It is about influence. The conscious mind cannot necessarily control everything (or much of anything). It can observe and it can influence.
For example, low-level inputs (e.g., sensory & external CNS inputs) to the brain are outside of the control of the brain. Low-level actuations of muscles, etc. can be well outside of the control of the conscious mind.
Even (more than likely) emergent mental entities like “thoughts”, “desires”, “dreams”, "feelings" etc. are not necessarily “controlled” (or necessarily even highly-influenced) by the conscious mind.
However, the conscious mind is quite clearly capable of observing all of these things (actions, colors, sensory inputs, feelings, etc.). Where “observation” is defined from the collectively-agreed upon (simple) description of our experience as human beings i.e. “the mind’s eye”.
Then we can argue that the conscious mind is some emergent "program" which accepts as input all of the information it can observe, and makes “decisions” based on this information likely alongside other internal information like cached calculations on prior observations (desires, beliefs), some sub-conscious neurological space “searches” (e.g., “critical thinking”/“reasoning”), and perhaps taking into consideration global chemical states ("feelings"), etc. and then attempts to have input (or influence) on the entities that it has observed e.g., actions, thoughts, desires, feelings, etc.
Is it always successful? No. Because, it is not about control. It is about influence.
This observer and it’s purely mental “actions” is the entirety of “free will”. A will but not necessarily a way.
Now, the question becomes: how is this internal (purely mental and emergent) observer able to attempt to have “input” on physical (arms, vocal chords, etc.) and mental (thoughts, feelings) entities?
My idea is that it doesn’t. It has “input” on emergent “representations” that are neurologically-stored discretely yet only exist in a spatiotemporal sequence of neural activity.
How could you have influence on something that is not a piece of memory, but rather a spatiotemporal sequence? If I change something stored in a single neuron then that will have profound effects on all possible spatiotemporal sequences that incorporate that neuron. Meaning, that a small change in the brain can never be a “small” change. Of course, certain spatiotemporal sequences may be robust/invariant to small modifications/perturbations but that only adds complexity to the outcomes of modifications.
The spatiotemporal sequence of neurons that are the emergent observer may also not always be active. In fact, there is a hypothesis floating around that the conscious mind is not always "turned on" during all phases of sleep.
This emergent program must also have the property of self-awareness. Our emergent observer is aware that it is aware. At least, this is often called “self-awareness”. Which we will define as the ability of our emergent observer to be able to observe at least some “part” of itself.
So, lets' take these facts into consideration when attempting to theorize a mechanism for this emergent observer exerting influence.
If “free will” is a result from an emergent observer observing all sorts of neurological activity, then how would it have “influence” on the system? My guess is that it does not necessarily have any input on the system external from itself. My guess is that it only has the option of self-modification such that when other neurological activity, outside of its control, interacts with it then those emergent entities will be modified as well.
In other words:
- the emergent observer program can only “observe” neurological activity that is actively interacting with one of "its " own neurons (or various other biological components that it may be emergent from). And this interaction must occur in a way that the program that is our conscious mind can undergo a perturbation by which to react to.
- It can only modify any of the neurons (or other components) within itself (as an emergent mechanism). This implies that the program can run purposeful neural modifications, in contrast to the majority of neural plasticity which is likely only induced from external low-level inputs. It purposefully makes neurological modifications likely by generating "artificial" low-level inputs (stimulating or depleting) by redirecting it's own execution resources — this redirection mechanism being the core "actuator logic" of the program. This also implies that the program must be insanely robust/invariant to modifications (which makes sense if it emerged from neural plasticity). One of the ways this could be accomplished is by a program that is highly polymorphic with respect to neural plasticity. Meaning that variations on the program still allow it to run within the bounds of some limited distribution of natural neurological changes.
- Thus, It expresses/performs influence by causing modifications to one of its own neurons that are directly used within some subset of the other emergent neurological programs that interact with it. In other words, the same set of neurons (or other biological entities) which act as the 'observation' neurons in the conscious mind are likely the same set of neurons that need to be purposefully modified to exert influence on the same process that was observed. In doing so, the external neurological process can itself be modified. This means that there is a class of neurological processes that never interact with the conscious mind directly: these are the sub-conscious neurological programs. The fact that sub-conscious neurological programs exist implies that the conscious mind is not emergent of the entire brain, and is only emergent from some subset of the brain.
In this way, the emergent observer can have profound influence (but not “control”) over any neural activity that it has direct contact with, including and especially other higher-level emergent neurological activity.
The little things that the human body does, like moving a single finger, it will not have much if any influence on. Finger movement is something that is left up to muscle neurons (“muscle memory”). But a much higher-level neurological phenomenon like: moving many fingers to play a piano or guitar beautifully, or planning piano and guitar practice materials in line with a self-improvement objective function, that it absolutely can have tremendous influence on. And since these higher-level emergent mental entities tend to have "supervisory"-level management on lower-level neurological activations (movement, etc.) then the emergent observer program can indirectly influence the entire body by only attempting to influence other higher-level emergent neurological programs.
This also explains other phenomenon like why a baby takes time to develop physical control over its body even though it is conscious.
The most interesting part of this hypothesized mechanism is the necessity for self-modification. One of the requirements of self-modification is self-reference (the ability to point to yourself). The simplest computational example of self-reference is recursion. But, this emergent observer must have the ability to reference certain parts of itself and modify those parts, and have all of that ability contained entirely within itself. Of course, this is not new to biology. Cells do this all the time. In fact, the basic premise of life itself could be based on the concept of self-modification.
Whereas in computational theory and mathematics we run away from “paradox” and “undecidability”, it is biology generally, and thus also this emergent observer program (i.e. the "conscious mind"), that embraces it in order to have persistence (and robustness) at all.
Self-reference is the basis for self-awareness. It shows that any emergent program that can achieve self-modification is an emergent observer program. But not all emergent observers have other higher-level emergent programs to react to and interact with e.g., "thoughts", "feelings", "desires", "emotions", etc. Some do. Like some animals. But most known self-referencing programs do not. So, then am I saying that recursive functions are self-aware? Yes. They are "aware" of something like some internal state (and sadly even they can "die" on exit conditions). But that is less self-"awareness" than a virus (or even a computer virus). I don't think we have any reason to worry about them "feeling pain" or expressing an existential crisis anytime soon.
We can test for the existence and location of the emergent observer in the brain (likely using existing Magnetoencephalography technology):
- Assume either (a) the emergent observer of the mind is not necessarily “on”/active during all parts of sleep (b) assume that the emergent observer is always active; Given either assumption (a) or (b) we can still search for a subset of the mind that meets one or the other criterion. My personal hypothesis about this emergent observer is that it is not a localized singular region of the brain (although it very well could be), but rather it is a subsection of each brain region (visual cortex, audio cortex, etc.). The rule would be that the emergent observer is active as long as any one of these subsections of the brain regions is active. In other words, the emergent observer requires temporal continuity on top of spatial continuity in order to be considered “on”. This multi-regional, temporally-continuous description would likely be true for other emergent cognitions such as “thoughts”, “feelings”, etc. The difference is that thoughts and feelings would not be persistent throughout being awake, and at least partially consistent throughout sleep. A study on many different participants could allow us to verify if this is true or not. Of course, if we work with assumption (a) then we also must be able to explain the mechanism of how the emergent observer program goes into a "sleep state" and then reawakens (and if being in a "coma" is some sort of reawakening bug in the program - which would also answer if coma patients are conscious and that answer would be "no").
- If experiment 1 fails to find anything then this could suggest that neural activity is not the foundation of the emergent observer and may be in fact require activity from multiple cell types e.g., glial cells + neural cells + other cells. And it is also possible that it is emergent from multiple systems e.g., nervous system + cardiovascular system + lymphatic system + other systems. This would be much more difficult to test for, but could still be done by overlapping many different system test’s data and looking for something that shows spatiotemporal continuity in satisfaction with either assumption (a) or (b).
- If no physical candidate is found to satisfy this hypothesis, then I will hypothesize that this would be the greatest evidence for an extended model of physics being required in order to explain certain self-"awareness" phenomenon like a conscious mind. This final hypothesis is third on my list as I see the least likelihood for it's necessity. But, human intuition is not always the best.