Choosing the Future: A Scientific Method

On a Saturday morning, with the window half-open and tea cooling, I picked up a quantum mechanics textbook I hadn't touched in years. Funny thing — the part that used to make my head spin now reads like an answer to a question I hadn't framed yet. Is the future already decided? And if it isn't, what exactly am I doing when I 'choose' it?

This column is an attempt to answer that — using only what physics, neuroscience, and cognitive science actually say. No mysticism layered on top. Just the structure as it stands, and the practice that follows from it.

01

Superposition is empirical fact

Run the double-slit experiment. Fire one electron at a time. With no measurement at the slits, the electron behaves as if it passed through both — building an interference pattern over thousands of shots. Place a detector at the slits, and the pattern collapses into two clean bands.

This is not philosophy. This is reproducible on any university lab bench. Until observation, the electron is not in a place — it is a spread of possibilities. The math that describes this spread is the wavefunction. The act of measurement turns the spread into one outcome.

02

Superposition is not the same as parallel worlds

Here a confusion has spread, and it is worth untangling. 'Superposition' and 'parallel worlds' come from two different interpretations of the same physics.

Same equations. Same experimental data. Two stories. Neither has been scientifically refuted. Saying 'quantum superposition equals parallel worlds' compresses two distinct interpretations into one, and the compression hides what is actually going on.

03

Where 'consciousness' has a legitimate seat

There is one more interpretation that should not be dismissed. John von Neumann (1932) and later Eugene Wigner (1961) argued that the only place in the chain of measurement that does not itself need a further observer is the observer's consciousness. So perhaps it is consciousness that does the collapsing.

Both were giants — von Neumann founded modern computing, Wigner won the Nobel Prize in 1963. Their hypothesis remained a minority view for decades, partly because it sounded mystical. But it never lost its formal standing.

In 2018, the Frauchiger–Renner theorem revived the debate with a fresh edge: it showed that quantum theory, applied consistently to observers themselves, leads to logical contradictions unless one of several deep assumptions is dropped. The role of the observer is once again on the table at the level of mathematical physics.

04

What is empirically established: observation shapes what you see

Step out of physics now. The next claim does not require any quantum interpretation at all. It is psychology and neuroscience, measured in controlled experiments thousands of times.

05

The practice in three steps

Put the four sections above together. Even if we set aside every speculative interpretation of quantum mechanics, what remains is robust: the brain is a prediction machine, attention is the steering wheel, and what you treat as the live possibility tends to become the lived one. The following practice is not magic. It is a deliberate use of well-documented mechanisms.

06

The trap of self-affirmation

Everything written so far can be hijacked by a single shortcut: repeating 'I am the best, I am the strongest' until the predictive model stops checking against reality. That isn't choosing a future. That is constructing a hallucination and calling it confidence.

Carol Dweck's two decades of work on mindset (Dweck, 2006) documents the difference. Praise and self-talk that anchors to identity — 'I am smart, I am gifted, I am the kind of person who wins' — produces fragile motivation. The moment the world pushes back, identity is at stake, and the brain defends rather than learns. By contrast, self-talk that anchors to process — 'I work through hard things, I update when I am wrong, I keep going' — produces the resilient kind of belief.

There is a darker version of the same trap. Mark Leary's work on hyper-egoism (Leary, 2007) and the wider literature on narcissistic self-enhancement show that unconditional positive self-talk, when never tested against feedback, produces inflated self-perception that is brittle, hostile to criticism, and predictive of long-run failure — not success. The mechanism is the same as before: a predictive model that does not update is no longer a model. It is a closed loop.

07

Metacognition — the observer observed

The missing piece between honest self-belief and self-delusion is one word: metacognition. John Flavell named it in 1979 — thinking about one's own thinking. Forty years of empirical work since has been consistent: people who can step back and watch their own predictions, beliefs, and emotional reactions in motion are better calibrated, learn faster, and recover more quickly from being wrong.

There is also a famous null finding: the Dunning–Kruger effect (1999). People who lack a skill systematically over-estimate how well they have it. The mechanism is metacognitive: the same gap that prevents them from being good at the task also prevents them from accurately assessing their performance on it. Self-belief without metacognition is structurally indistinguishable from this gap.

What this looks like in practice: alongside 'I am orienting toward this version of myself,' ask, every day, 'what am I currently believing about myself that might not be true?' The first sentence sets direction. The second one keeps the model honest. Drop the second, and direction-setting becomes a closed loop. Keep both, and what you are doing is not affirmation — it is calibrated steering.

08

An image worth holding: the mahjong table

Strong mahjong players know the gesture without needing any of the science above. The whole structure of this column is already encoded in how a competent player sits at the table.

But the deepest thing the mahjong table teaches is the move the first seven sections only hinted at.

This is what 'choosing the future' actually means at the human scale. It is not the rigging of a single outcome. It is the deliberate biasing of the long-run distribution by repeated calibrated action. The single hand belongs to chance. The probability distribution belongs to you.

Two phrasings of the same idea, one weak and one strong, often get confused. 'If I act with attention, I will win' is the weak one — it breaks on the first lost hand and tempts the player into either denial or despair. 'If I act with attention consistently, the distribution of my outcomes will shift toward winning' is the strong one, and it survives the lost hands because it never promised to avoid them.

So yes. Believing in yourself works. Not as a guarantee for the next round. As gradient on the cumulative one. The single hand belongs to chance. The aggregate — the version of yourself that emerges after hundreds of calibrated rounds — that one is being chosen, hand by hand, by what you do when you sit down at the table.

09

Beyond the model: the territory of tacit intuition

Everything written so far is explicit. Equations, references, predictive coding, mahjong gestures — all of it can be put into words and taught. And yet anyone who has actually run a hard decision in the real world knows: the move that turns out to be right is rarely the one you can fully justify at the moment you make it. Something else is operating.

Michael Polanyi named this layer in 1966 with a single sentence: 'we can know more than we can tell.' He called it tacit knowledge. The chess grandmaster who 'sees' the best move without computing variations, the experienced doctor who 'senses' a diagnosis before the tests come back, the founder who 'knows' a market is about to shift — all of them are running on tacit knowledge, and none of them can fully explain to a beginner how they did it.

Paul Dirac, one of the founders of quantum mechanics, said it from the inside of physics: choose the beautiful equation, and if it doesn't match experiment, suspect the experiment first. This is not whim. This is forty years of immersion in mathematical structure, compressed into an aesthetic sense that points faster than calculation. Beauty, for Dirac, was a working instrument — not decoration.

The same phenomenon appears in business. CyberAgent founder Susumu Fujita is widely cited in Japan as the executive who pivoted his entire company to smartphone-first around 2010 — well before the market data justified it. Asked in interviews how he knew, Fujita has consistently given a non-explanation: 'it was going to come.' Not a model. A body-level certainty arrived at after a decade of full immersion in the internet industry. The intuition was the compressed product of that decade.

So the full picture, with all nine sections on the table, looks like this. Attention sets the direction. Predictive coding builds and updates the model. Metacognition keeps the model honest. The mahjong table teaches that the long-run distribution is what you are actually steering. And on top of all of that — only on top of all of that — sits the layer where tacit intuition begins to operate: the layer where, after enough immersion, the body starts knowing things faster than the model can compute them.

This is the layer that the most rigorous people in any field eventually point to with the most cautious words. The physicist says 'beautiful.' The doctor says 'a feeling.' The mahjong player says 'kuru' (it's coming). The founder says 'I knew.' It is the same layer. And the path to it is never a shortcut — it is the cumulative work of the eight sections above, repeated until the work disappears into the body.

Coda

So what is 'believing in yourself,' precisely?

It is not a feeling, though it shows up as one. It is three acts held together. The first: orienting attention, prediction, and behavior toward a particular version of yourself, consistently enough that the salience filter, the predictive model, and the day's choices begin to align. The second: stepping back periodically to ask whether the model itself still matches reality, and updating where it does not. The third: staying in this work long enough — and deeply enough — that the work begins to disappear into the body, and what remains is the layer Polanyi pointed to. The layer that knows faster than thinking.

The first without the second is self-hypnosis. The second without the first is paralysis. The third without the first two is impossible — there is nothing to compress. All three together form what could be called calibrated steering ripened into tacit knowing — the version of self-belief that survives contact with reality, because reality has been treated as an ally for long enough that the body finally trusts it.

The textbook went back on the shelf. The tea was cold by then. The question that opened this column — is the future already decided? — turns out to have a strangely precise answer. Not entirely. Not yet. And the part that is not yet decided is the part you can work with, provided you are willing to keep checking which part that actually is.