The Brain's War on Uncertainty

There is a question Isaac Asimov asked, and it has never really left me.

In his short story called The Last Question, humanity builds increasingly powerful computers and keeps returning to the same urgent problem across billions of years: can entropy be reversed? Can the universe's inevitable drift toward disorder be stopped, or even turned around? The computers keep answering the same way — not yet, insufficient data — until, at the very end of time itself, one of them finally figures it out.

Most people read that story as science fiction. I kept reading it as something else entirely: a portrait of the human mind.

Because what if evolution — not the slow biological kind, but the kind we are capable of right now, in a single lifetime — is fundamentally about this same question? Not how we find better answers, but how we learn to live more honestly inside the uncertainty while we wait for them?

The Brain Is an Entropy-Fighting Machine

Here is something that neuroscience has confirmed, and that most of us experience without ever naming it: the brain does not passively receive reality. It predicts it.

According to the framework known as Predictive Processing — developed and refined by researchers like Karl Friston and Andy Clark — the brain is constantly generating a model of what is about to happen. Before your eyes finish processing the shape in front of you, before the words in a sentence are complete, your brain has already made a guess. It is not a passive recorder. It is an active forecasting system, running perpetually against reality and quietly updating when its predictions are wrong.

The goal of this system, at its most fundamental level, is the minimization of surprise. Prediction error — the gap between what the brain expected and what actually happened — creates a kind of neural discomfort. Not just discomfort: threat. When reality consistently fails to match our model, the nervous system responds as though something is wrong. Because for most of human history, something was.

This is the brain's version of fighting entropy. Every prediction is an attempt to impose order on chaos. Every mental model is a fortress built against the unknown.

And it works extraordinarily well — until it doesn't.

The Default Mode and the Trap of the Familiar

There is a network in the brain that activates most powerfully when we are not focused on the outside world — when we are daydreaming, ruminating, imagining the future, or processing the past. It is called the Default Mode Network, and it is one of the most energy-intensive systems in the human brain.

What it mostly does is tell us stories about ourselves. It constructs and maintains the narrative of who we are, what we expect, what we deserve, and what tends to happen to people like us. It is, in a very real sense, the machinery of identity.

This is not a flaw. It is an elegant solution to the problem of living inside a complex social world. We cannot evaluate every situation from scratch. We need templates, shortcuts, running assumptions about how things work. The Default Mode Network provides them.

But there is a cost. The same system that stabilizes our sense of reality also limits it. When a new situation arrives that does not fit the model, the brain's first instinct is not to update the model. It is to force the situation into the model's existing categories. To recycle the past in order to manage the present.

You have almost certainly felt this. The conversation that triggers an old wound and suddenly you are not responding to what is happening now but to everything that happened before. The new opportunity that gets quietly filtered out because it doesn't match the shape of what you've believed was possible for you. The person in front of you who never quite gets seen — because what you are really looking at is a projection assembled from everyone who came before them.

The Default Mode Network is running, and it is efficient, and it is keeping you safe from a certain kind of uncertainty. But it is also keeping you stuck inside a loop.

What Real Evolution Actually Requires

We use the word evolution loosely — to mean growth, improvement, becoming. But if we take the neuroscience seriously, evolution in any meaningful sense is not primarily about accumulating better answers. It is about developing a higher tolerance for not knowing.

A system that can only function when its predictions are confirmed is a brittle system. It can perform flawlessly in familiar territory and collapse entirely when something genuinely new appears. What distinguishes a system capable of real development is not the quality of its current models, but its capacity to hold uncertainty without immediately collapsing back into old ones.

This is what psychologists call distress tolerance — the ability to remain functional and curious in the presence of ambiguity, rather than reaching for the nearest available certainty. It is not a passive state. It is an active, effortful discipline. And it is one of the rarest capacities in human life.

Because the nervous system does not naturally want to sit with open questions. It wants resolution. It wants the entropy resolved, the gap closed, the uncertainty converted into something manageable and known. Every anxious thought, every compulsive need to settle a question before it is ready to be settled, every snap judgment made to end the discomfort of not knowing — all of it is the brain doing exactly what it was designed to do. Trying to reverse entropy.

To resist that pull — to stay with the question — is to go against a very deep biological grain.

The Connection to How We Work Together

This is not only a personal challenge. It is a collective one — and it is, we would argue, the central challenge of collaboration in complex, fast-moving environments.

Most organizational dysfunction is not caused by a lack of intelligence or effort. It is caused by an excess of certainty. By teams that have converged prematurely on a way of seeing the problem. By leaders whose models of their industry, their people, or their strategy have stopped updating. By groups in which the social cost of introducing uncertainty — of saying: “I don’t know” or “What if we’re wrong?” — is too high to bear.

What entropy-intolerant teams do — consciously or not — is optimize for resolution. They move quickly to consensus not because they have genuinely aligned, but because the discomfort of open disagreement is too uncomfortable to sustain. They make decisions before the right questions have been fully explored. They mistake the quieting of uncertainty for the presence of clarity.

The teams that consistently outperform — the ones that can navigate genuine complexity, adapt to real disruption, find solutions that weren't obvious at the outset — are typically not the most brilliant. They are the most tolerant of productive tension. They can hold multiple competing possibilities at once. They can sit in the space between knowing and not knowing without rushing to collapse it.

That capacity doesn't emerge by accident. It is cultivated through specific practices: creating psychological safety for honest dissent, developing the skill of asking rather than asserting, building the kind of trust in which uncertainty can be expressed without triggering defensiveness. It requires people who have done enough of their own inner work that they can distinguish between a genuine open question and a threat to their identity.

The Invitation at the Edge of the Unknown

There is something quietly radical about Asimov's premise — that the most important question humanity ever asks is one that takes the entire lifespan of the universe to answer.

We live in a moment that is deeply uncomfortable with that pace. Everything in the current environment rewards rapid resolution. Fast takes, quick decisions, efficient processing, the confident assertion that cuts through ambiguity. We have built systems — technological, organizational, cultural — that are extraordinarily good at generating answers and extraordinarily bad at evaluating whether we are asking the right questions.

The rise of artificial intelligence is, among other things, an exponential acceleration of answer-generation. Machines are becoming astonishing at producing outputs — content, predictions, recommendations, analysis — at a scale and speed no human can match. What they cannot do, at least not yet, is remain genuinely open to a question they don't yet know how to frame.

That, perhaps, is where the human edge lives. Not in being faster than machines at producing certainty, but in being wiser about when certainty is premature. In the capacity to ask a better question. To stay with something unresolved long enough for it to reveal what it actually is.

This is not passivity. It is a specific, disciplined, and deeply human kind of courage. The courage to expand your internal model rather than defend it. To tolerate the noise of not knowing long enough for something genuinely new to become clear. To stop defaulting to automatic predictions and begin, however slowly, to live in a wider space.

At SavvySimian, we work with people and teams who are trying to do exactly this — not because it is comfortable, but because it is necessary. The collaboration challenges that matter most right now cannot be resolved by better systems or smarter processes alone. They require human beings who have developed the capacity to hold complexity without collapsing it, to stay curious in the presence of what they don't yet know, and to invite each other into that same uncertain, generative space.

That is not a small thing. It may, in fact, be the whole thing.

Evolution is no longer about better answers. It is about increasing the capacity to stay with the question.