Free Will as an Experience

Free will is an experience. A system that computes projections of possible outcomes, but enters a threshold where several options carry a similarly high probability, experiences free will as its computation starts to loop.

Let's look at free will from a few different perspectives.

Free will as qualia

Consider the last difficult decision you made. That feeling of deliberation, of being poised between options, is free will as a sensation. It is not fundamentally different from other qualia like the warmth of sunlight on your skin, the redness of an apple, or the emotion of love.

If we accept this, a profound realization follows: every qualia can be mapped into a computational information matrix. Just as the qualia of pain can be replicated in computation (think of a game engine where damage inputs trigger a "pain state" that forces behavioral changes in an NPC), the specific sensation of free will must also be extractable. Free will is what it feels like when a brain processes possible outcomes and computes which direction to take based on finer and finer variables. It is deterministic, and yet from the perspective of consciousness it is a legitimate, computable sensation of freedom that we should be able to isolate, even if we have not managed to yet.

Notice also that most computations are replicas of sensations. We needed pain and free will first in order to mimic them in computation. The meaning, and the depth of what is computable, is already inherent in perception. We just extract it.

The strange loop that creates free will

Our brains are prediction machines caught in a strange loop: they predict their own predictions. When you make a choice, your brain computes what you will decide, but that prediction itself becomes a factor in the decision, which changes the prediction, which changes the decision, creating infinite recursion.

Think of a system predicting its own coin toss. If it predicts "heads", that prediction pushes the outcome toward "tails" to prove itself wrong. But anticipating this self-falsification pushes it back toward "heads". This recursive chase, where each prediction must account for itself, generates the sensation of free will. The more balanced the options, the deeper the loop runs, and the stronger the feeling of choice becomes.

The perspective paradox

Reality is always experienced from a single perspective, and this creates an unsolvable puzzle for free will. Imagine John observing Jane's neural activity to predict her choices. Jane appears deterministic to John, but John experiences free will. Add a third observer watching John, and now John appears deterministic too. Add a fourth, a fifth, an infinite chain of observers.

Here is the crucial insight: no matter how many observers you stack, there is always one final perspective - yours, reading this - from which free will can be neither confirmed nor denied. Like quantum collapse, where observation happens from a single viewpoint regardless of the chain of observers, free will lives in this irreducible blind spot of consciousness. You can never step outside your own perspective to see whether your choices are truly free or merely feel free. This is not a limitation of knowledge. It is the structure of reality itself. There will always be exactly one perspective, the one experiencing reality right now, where the question stays open.

The prediction race

Imagine a machine that scans your brain to predict your choice one second before you make it. In actual experiments where the machine keeps its prediction hidden, success rates are surprisingly low, barely above chance, which suggests our neural complexity already resists simple prediction. But reveal the prediction ("you will choose red") and the system breaks entirely. Now your mind can defy it and choose blue. The machine scans your defiance and updates its prediction, which you can defy again, and so on forever.

This reveals something important. Even without knowing the prediction, our brains are already too complex for current machines to predict reliably. Add the feedback loop of a revealed prediction, and if consciousness has any non-physical component, the chase becomes impossible. The game of cat and mouse drives deeper into the brain's physical substrate, from neural networks to atoms, and finally into quantum mechanics. The prediction would need ever finer resolution, ultimately demanding a perfect replica of your brain.

But here is the paradox: a perfect physical replica would compute at the same speed as your actual brain, making its "prediction" simultaneous with your decision rather than ahead of it. At this boundary the question becomes unanswerable, not through lack of knowledge, but because the measuring tool has become indistinguishable from the thing it measures. Hidden predictions barely work. Revealed predictions create paradox. The gap where prediction fails might be where free will lives.

Perhaps this is what free will really is: not the absence of causation, but causation so complex, so self-referential, so responsive to its own observation that it generates genuine novelty. A strange loop where cause and effect chase each other in circles until they blur.

The pragmatic proof: blame

Step outside the brain's internal loops and there is a functional definition of free will built into our social dynamics. Consider blame. Society assigns ownership of actions to individuals and blames them for undesirable outcomes. Crucially, this input changes future behavior.

The person receives an external attribution of responsibility, processes the negative feedback, and recalibrates their future choices to avoid similar outcomes. In this pragmatic view, free will is real simply because blame works. It is a functional feedback loop. Any system that can receive blame (data about ownership of a past error) and use it to change its own future behavior can effectively be said to have free will.

Mapping free will through pattern recognition

Let's assume free will exists as something beyond physical causation, something almost magical. Even then, if we could rewind time and replay a decision, we would likely see the same choice made again. Why? Because a specific Mind in a specific body, in a particular mood on a particular day, facing specific options, would exercise its free will in a characteristic way. That consistency, whatever its source (physical, spiritual or otherwise), is data we can record.

Imagine a vast dataset: millions of Minds in different bodies, moods and contexts, all making free decisions. We are not tracking causation, only recording how free will shows up across countless situations. With enough data, patterns would emerge. Not patterns of determinism, but patterns of how this mysterious property tends to express itself under certain conditions. We could then train neural networks not to understand the mechanism of free will, but to predict its behavior through pure pattern recognition, the way we can map the wind without understanding aerodynamics.

The irony is that we might predict free will's behavior without ever proving whether it is truly free, turning the ineffable into data points while keeping its essential mystery intact.