Your brain has already decided what it is looking at before you know you are looking at it. Not as a metaphor. Not as a figure of speech. As a measurable, mathematically describable physical process that happens in the gap between a signal arriving and a thought forming, a gap so small you will never feel it, but large enough for everything you believe to pour through it and reshape what comes out the other side.
Findings published in Scientific Reports in April 2026 map a framework in which the observer’s mental state, their beliefs, emotional orientation, and degree of skepticism, physically alter incoming sensory data before conscious classification occurs. Note: this is a preprint and final peer review editing is ongoing.
Start with something you think you know. You are looking at an object. Light hits your retina. Signals travel to your brain. Your brain identifies the object. That sequence feels mechanical, neutral, faithful to whatever is actually there. It is not. Between the signal arriving and the identification landing, your brain runs the incoming data through everything it already believes, and that process is not passive filtering. It is active rewriting. The signal that enters the classification stage is not the signal that arrived at your eyes. It has been changed. The change is not random. It is shaped, directionally, by who you are and what you expect.
The math that describes this most accurately turns out to be quantum math. Not because the brain is doing quantum physics at the neuronal level. It is not. But because quantum probability, the branch of mathematics built to handle systems that exist in multiple possible states simultaneously and collapse into a single outcome only at the moment of measurement, maps onto what perception actually does more accurately than classical probability does. Classical probability assumes the thing being measured has a definite state before you look. Quantum probability does not. It treats the state as genuinely unresolved until the measurement happens. Perception, it turns out, works the same way. What you see is not retrieved from reality. It is produced at the moment of looking, and the production process is shaped by the instrument doing the looking, which is you.
The interaction between incoming sensory data and the observer’s mental state is modelled using an equation physicists use to track how a quantum particle behaves when it is weakly coupled to its surrounding environment. The sensory data plays the role of the particle. Your mental state plays the role of the environment. The two interact before classification begins. The strength of that interaction depends on how strongly your current beliefs couple to the incoming signal, and on how long the interaction runs before a decision is made. Both of those parameters are, in principle, experimentally measurable. One of them is directly analogous to reaction time. The longer the gap between receiving information and acting on it, the more your mental state has reshaped the signal before you respond.
This has a specific and uncomfortable consequence. Two people in the same room, looking at the same thing, are not running the same perceptual process. They are not receiving the same signal and interpreting it differently. They are receiving the same raw input and converting it into genuinely different signals before interpretation begins. The classification step, the moment where the brain decides what category the thing belongs to, is mathematically distinct for each observer because it is constructed from that observer’s mental state at the moment of measurement. Change the mental state, change the measurement process, change what gets seen. This is not opinion. In the framework, it is a parameter with a value.
The clearest demonstration of what this means in practice comes from a simulation. An artificial observer is shown a stimulus that is clearly defined across its top half but carries zero information about its bottom half. The straightforward result would be a clean classification matching the top half only. Instead, the observer’s orientation toward trust and acceptance drives it to prefer a two-part classification that includes a specific bottom-half feature, despite no bottom-half data existing in the input. That preference scores a 10 percent probability. The simpler, better-supported classification scores 4 percent. The wrong answer, by any evidential standard, wins because the observer’s anticipated mental state after a correct classification, meaning the emotional posture the brain expects to land in once it has successfully identified something, pulls the outcome toward the category that produces that posture. The brain is not finding the best match for the data. It is finding the match that lands it where it wants to be.
Skeptics and believers do not just disagree. They process information through formally different mechanisms. A believer, in this framework, runs a tight classification process. Incoming features are matched strictly against existing categories. Anything that does not fit is assigned a near-zero probability of being real. The outcome is sharp and confident, but it is also brittle. Information that falls outside existing templates gets cut. A skeptic runs a broader process. Unknown features are not dismissed. They are assigned a flat distribution across all possible categories, meaning the skeptic treats gaps in their knowledge as genuinely open rather than closed. The outcome is noisier and less decisive, but it catches things the believer misses. Neither posture is more accurate in absolute terms. Each is optimised for a different relationship with uncertainty, and that optimisation is encoded in the measurement process itself, not bolted on afterward as a cognitive style.
What grounds this in something physical rather than purely theoretical is a body of neuroscience findings the framework draws on directly. Brain state before a stimulus arrives shapes how that stimulus gets processed, a documented and replicated effect across multiple experimental paradigms. When a classification is genuinely ambiguous, the brain produces a specific electrical signature, a late positive potential detectable by EEG, that is distinct from the signature of low signal strength. Ambiguity is not just weak certainty. It is a different state entirely, represented differently in the brain. Sensory uncertainty is encoded in the visual cortex as probability distributions over possible stimulus orientations, not as single best-guess values. The parietal cortex carries a direct causal role in computing subjective probability estimates when inputs are ambiguous. Every one of these findings maps onto a specific component of the framework. The math is not floating free of biology. It has anchors.
The gap between where this sits now and where it needs to go to become testable science is real and the researchers do not obscure it. The framework contains parameters, including observer temperature, oscillator frequencies, and coupling constants, that currently have no direct experimental mapping to measurable psychophysical variables. Temperature in this context describes how broadly or narrowly an observer’s prior beliefs are distributed across possible states. A high-temperature observer carries diffuse, non-committal priors. A low-temperature observer is tightly committed. These are formally meaningful distinctions, but converting them into numbers that can be read off a real human subject requires experimental tools and protocols that do not yet exist for this purpose. The simulation software is publicly archived. The parameter dataset is publicly archived. The next step is empirical work to test whether the model’s predictions match measured human behavior.
The preprint currently sits at accepted manuscript stage in Scientific Reports, published online April 14, 2026. Final editing is ongoing. What it proposes, if it survives that process and subsequent experimental scrutiny, is a framework in which subjectivity is not a flaw in the system of human perception. It is the system. The observer is not standing outside the measurement. The observer is built into the measurement at the level of the equations, with a coupling strength, a temperature, and a parameter value between zero and one that describes exactly how open or closed they are to what they do not already know. The question of whether two people can ever truly see the same thing gets, for the first time, a formal mathematical address.
Source:
Hoorn, J. F. & Ho, J. K. W. (2026). Observer effect modulates classification in a quantum epistemic framework. Scientific Reports. https://doi.org/10.1038/s41598-026-46604-9
https://www.nature.com/articles/s41598-026-46604-9
