Here is something most people find unsettling when they first encounter it: you are not seeing reality. You are seeing your brain’s prediction of reality, updated by incoming evidence only to the degree that the evidence is sufficiently different from what was expected. It is the current consensus in neuroscience, and it has direct consequences for every market assessment, customer conversation, and competitive analysis you have ever done.

Your brain is a prediction machine, not a camera

Karl Friston’s predictive processing framework — one of the most influential theories in contemporary neuroscience, though contested in its stronger philosophical claims — reverses the naive understanding of perception. The traditional view: sensory information arrives, the brain processes it, and you experience reality. The actual view: your brain generates a continuous prediction of what it expects to receive, compares incoming sensory data against that prediction, and only updates the prediction when the mismatch is large enough to warrant it.

Perception is predominantly top-down rather than bottom-up. What you experience is a construction — a sophisticated, continuously updated prediction generated from prior experience, context, and the small proportion of incoming data that sufficiently violates the existing model.

The cost of this efficiency is that the system jumps to conclusions based on partial information and is sometimes wrong. The benefit is that it is metabolically cheap and extremely fast. Evolution optimised for speed, not accuracy. Your brain is very good at its job. Its job is not showing you reality.

The practical implication: what a founder experiences as “seeing the market” is largely “seeing their prediction of the market.” The customer conversations, the competitive signals, the market data — all of it is filtered through and partly constructed by prior beliefs. A founder with seven years in a space and a founder entering it for the first time will construct genuinely different perceptual realities from identical information. The experienced founder knows more. They also see less of what is actually there.

Why the brain acts to confirm its predictions rather than just receive information

Predictive processing theory identifies two strategies the brain uses to reduce prediction error — the gap between what it expected and what arrived. The first is updating the prediction to fit the data. The second is acting on the world to make the data fit the prediction.

That second strategy is the neurobiological basis of confirmation bias — and it is not a flaw. It is the system operating efficiently. A founder who believes their product solves a real problem will conduct customer discovery in ways that make confirming responses more likely. Through the questions they ask, the way they frame them, the signals they respond to, and the responses they record. They are not being dishonest. Their nervous system is actively generating sensory consequences — customer responses — consistent with its prediction.

Friston and colleagues’ active inference framework, published in Neuroscience and Biobehavioral Reviews in 2017, formalised this: the brain does not merely observe the world. It acts on the world to produce the observations it expects. For an entrepreneur validating a hypothesis, this means the validation process itself is partly shaped by the hypothesis being tested.

The practical correction is structural rather than intentional. “How valuable would it be if X?” is active inference — it primes a positive response. “What do you do when you have this problem, and how much does it bother you?” is closer to genuine inquiry. The difference feels small. The data it produces is significantly different.

Why strongly held beliefs resist updating

The third mechanism explains something every entrepreneur has observed without necessarily having a name for it: a founder who has been in a market for years can hear consistent negative signals and remain genuinely unconvinced.

The Bayesian brain hypothesis — related to but distinct from predictive processing — describes how the brain weights incoming evidence against prior beliefs. The weight assigned to incoming data relative to the prior depends on each one’s estimated reliability, or precision. When a founder has a high-precision prior — a strongly held, long-held, deeply experienced belief about their market — incoming evidence needs to substantially and consistently exceed that precision threshold before the prior updates.

A single piece of negative customer feedback does not meet that threshold. Ten pieces of inconsistent customer feedback do not meet it either if the prior is strong enough. The founder is not being irrational. They are operating a system that, by design, requires substantial disconfirmation before revising well-established beliefs.

This is why informal market proximity is insufficient for a founder with a strong prior. Being close to customers while filtering their feedback through a high-precision prior produces evidence that confirms the prediction. Formal, structured, high-quality disconfirmation — rigorous market research, independently conducted customer interviews, external advisory perspectives — is required to generate evidence at the precision level needed to exceed the prior’s threshold.

It is also why diverse teams genuinely see different things. Team members with different prior experiences construct different perceptual realities from the same market information. That is not a communication problem. It is the actual mechanism of how prediction-based perception works — and it is the closest thing to a structural solution available.

A book worth reading alongside this

Surfing Uncertainty by Andy Clark is the most accessible scholarly treatment of predictive processing and its implications for how the mind constructs reality. Clark is one of the primary philosophical architects of the framework, and the book translates the technical neuroscience into something readable without losing the rigour behind it. For any entrepreneur who wants to understand not just that they are constructing reality but how, it is the most intellectually honest starting point available.

This article discusses neuroscientific frameworks documented in research on predictive processing and perception. The predictive processing framework is influential and well-supported in its core empirical claims, though contested in its stronger philosophical versions. It is not designed to identify, diagnose, or assess any psychological condition. If patterns of belief rigidity or perception are significantly affecting your business decisions or relationships, speaking with a psychologist can provide personalised support.

This article is for educational and informational purposes only. It is not a substitute for professional psychological advice, diagnosis, or treatment. If you are experiencing significant psychological distress, please consult a qualified mental health professional.

Sources: Friston, K. (2010), Nature Reviews Neuroscience, 11(2), 127–138. Friston, K. et al. (2017), Neuroscience and Biobehavioral Reviews, 68. Hohwy, J. (2013), The Predictive Mind, Oxford University Press. Clark, A. (2016), Surfing Uncertainty, Oxford University Press.