Why the Most Experienced Leaders Are Often the Most Dangerously Wrong
- 2 days ago
- 7 min read
Sarah Merron is an NLP Trainer and Self-Leadership Coach who has transformed mindsets since 2008. With her extensive experience, she has trained and coached individuals from all walks of life, including business leaders, entrepreneurs, parents, athletes, and more. Sarah's expertise lies in Neuro-Linguistic Programming (NLP), Hypnosis & Time Line Therapy® techniques. For 15 years, she has been helping people develop actionable self-insight and self-awareness and map their version of success. With her guidance, you can design an authentic life that truly feels like you.
What predictive coding reveals about expertise, perception, and the quiet architecture of decision-making. There is a paradox at the center of senior leadership that almost nobody names out loud. The longer you lead, the more certain your decisions feel. The pattern recognition gets faster. You deliberate less. Years of hard experience build something that feels, from the inside, like reliable instinct. Like clarity, you have earned.
Neuroscience suggests that the same process might be building something else at the same time. A model of the world that gets harder to update the more successful you become. The confidence is real. The clarity is genuine. Under certain conditions, both quietly turn into a liability you cannot feel.
Your brain does not perceive, it predicts
Predictive coding is currently the leading account of how the brain builds your experience of reality. You are not a passive observer taking in the world and then working out what it means.
Your brain is constantly guessing what comes next. As new information arrives, it compares what it expected with what actually turned up. It also pays attention almost entirely to the gap between the two. Those gaps are called prediction errors. The predictions themselves are automatic, shaped by everything you have lived through before.
So what you experience in any given moment is not reality. It is your brain's best guess at reality, corrected only where the incoming data disagrees sharply enough to matter. Your brain does not process everything. It processes the surprising. The model fills in the rest. Which is fine, until the model gets very good, very fast, and is very rarely wrong.
What experience actually builds
Experience builds priors. A prior is just the brain's expectation before the new information lands. The stronger the prior, the more confidently the brain predicts, and the harder it pushes away anything that contradicts it.
A founder who has built and scaled a company has a rich, well-founded set of priors about markets, people, risk, and timing. Those priors are genuinely valuable. They are what let you read a room or a deal in seconds when someone less experienced would still be reaching for the data.
They also come with a specific trap. Stress pushes the brain to favour speed over accuracy when it forms its predictions, which makes it more likely to lock in a belief that is precise and confident, and also wrong. Stress is the water senior leaders swim in. Sustained pressure, high stakes, no time. Under exactly those conditions, the brain trades accuracy for speed, and hands you a prior held with total confidence, acted on fast, and potentially well off the mark.
You do not experience any of that as a shortcut. You experience it as decisiveness. That is where certainty turns dangerous.
Every belief your brain holds carries a weight, a sense of how much to trust that expectation against the evidence in front of you. When that weighting is calibrated well, contradictions get noticed, taken in, and the model continues to learn.
When a prior becomes too heavily weighted, something else happens. The brain rigidly holds the belief and quietly discounts anything that argues with it. The prediction errors are still there. They do not register, and the model stops updating.
From the inside, this does not feel like rigidity. It feels like experience. The information that contradicts you feels like noise, or hesitation, or someone junior failing to see the obvious.
At that point, your confidence is being generated by the model itself, not by an accurate reading of what is actually in front of you.
The brain does not stop at predicting. Through what is called active inference, it acts on the world to make its predictions come true. For a leader, that means your model quietly shapes strategy, hiring, structure, culture, and the whole information environment around you. Believe a market is mature, or that one approach to competition is right, and you will build the very structures that produce evidence for it. When the prior is correct, that is momentum. When it is wrong, active inference accelerates the damage rather than catching it.
The updating problem
A healthy predictive brain updates as new evidence arrives. However, not all evidence gets equal treatment. Anything that fits your current beliefs and mood gets waved through. Anything that contradicts your experience or arrives while you are under pressure gets quietly marked down.
Here is the strange part. This is not a flaw in the leader. It is a flaw in the room. Think about the last time someone told you something you did not want to hear, and told you straight, no softening, no run-up. If you are near the top of an organisation, you can probably think of fewer of those moments each year, not more. That is not an accident, and it is not because the people around you got worse at their jobs. It is because every layer between you and the truth has a reason to hand you a version that is easier to say.
The information gets smoothed. Quietly, without anyone deciding to do it. By the time something reaches your desk, it has already been rounded off by three or four people, each of whom has their own small reason to soften it.
Your brain does not know the difference. It is not built to. It just does what brains do. The brain builds a working model of how things are, updating that model whenever reality pushes back hard enough to register. That pushback is the whole mechanism. No pushback, no update. Not because you have got it right. Because there is nothing left to correct you.
This is the part that should unsettle people more than it does. Confidence and accuracy grow together early on, when the feedback is loud and unfiltered, and nobody is cushioning it for you. Somewhere on the way up, they quietly come apart. You do not feel it happening. Conviction feels the same whether it is earned or handed to you by a room that has stopped telling you the truth.
When the model stops updating: Nokia
In 2007, Nokia held more than 40 percent of the global mobile phone market. By 2012, that was gone. In 2013, its phone business was sold to Microsoft for a fraction of what it had once been worth.
The usual story is that Nokia failed to respond to the iPhone. That is true, but it is not the whole picture. The sharper version is that Nokia's leadership had built priors so reinforced by years of dominance that the organisation could no longer accommodate the corrective signals it was generating inside its own walls.
People who worked there later described what it actually looked like. Employees had spotted the threat. Concerns were being raised. But the environment around senior leadership had been shaped, over years of winning, so that dissenting signals were filtered out before they reached the people whose model most needed them. Middle managers held back the unwelcome assessments, not because they lacked the information, but because the cost of contradicting the prevailing view had quietly become too high.
The corrective signal existed. It was generated. It was suppressed before it could disturb the prior. Nokia's leaders were not stupid or incurious. They were sitting inside an information environment that their own success had slowly reshaped to confirm what they already believed. The model stopped updating. Not suddenly. Gradually, and then all at once.
What this actually asks of you
You cannot think your way out of this one. Insight does not fix it, because the problem was never inside your head. It is in what reaches your head. More information does not fix it either, because a brain with overweighted priors will discount the new information just as efficiently as the old.
What resolves it is building another information pathway that does not run through the usual people. Someone senior enough to disagree with you and pay no price. A way to hear from the front line before it has been translated for you. A habit of asking what you are not being told, and actually waiting for the answer.
Underneath that, there is a disposition worth building on purpose. The genuine appetite for evidence that contradicts your thinking. The willingness to sit in unfamiliar rooms where your priors do not work. This is not a soft skill. It is not open-mindedness. This mechanism is how a predictive brain stays calibrated to a world that keeps moving.
There is a caveat. Predictive coding is a neuroscientific framework, not a management method, and the leap from how the brain minimises error to what a leader should do on Monday morning passes through many intermediate steps.
However, the core of it stands. The leaders who most need to update their model are exactly the ones whose model is most confident, most rewarded by success, and most defended by the very certainty that got them here. In a predictive brain, certainty is not the destination. It is the warning sign.
Read more from Sarah Merron
Sarah Merron, NLP Trainer & Self-Leadership Coach
Sarah Merron is an NLP Trainer and Self-Leadership Coach who has transformed mindsets since 2008. With her extensive experience, she has trained and coached individuals from all walks of life, including business leaders, entrepreneurs, parents, athletes, and more. Sarah's expertise lies in Neuro-Linguistic Programming (NLP), Hypnosis & Time Line Therapy® techniques. For 15 years, she has been helping people develop actionable self-insight and self-awareness and map their version of success. With her guidance, you can design an authentic life that truly feels like you.
