H U M A N I N T E L L I G E N C E A R C H I T E C T U R E ™
Projection at Work
A factual remark lands as an attack.
A moved deadline reads as disrespect.
The mismatch is not oversensitivity — it is an old model firing, and the person in front of you is paying for it.
MELANIE ALBISSER · HUMAN INTELLIGENCE ARCHITECT · 6 MIN
Everyone has sat in a meeting where a reaction did not fit the situation. A factual piece of feedback lands as an attack. A rescheduled deadline is received as disrespect. A clarifying question is answered as if the whole person had been called into doubt. Everyone in the room registers the mismatch, and almost no one names it.
The explanation is not that people are oversensitive. The explanation sits in how the brain handles information, and in how little of that handling ever reaches awareness.
The part you never see
Your nervous system takes in a very large stream of sensory data every second. Only a tiny fraction of it becomes conscious experience. The exact ratio depends on how you measure it, and researchers argue about the numbers, but the order of magnitude is not in dispute: what you consciously notice is a rounding error compared to what your brain processes without you.
Everything else is handled by systems that never ask permission. Tone of voice, micro-expressions, posture shifts, the length of a pause before someone answers, the temperature of a room. All of it is read, matched against stored patterns, and turned into a body state and an impulse before the thinking part of your brain has been informed that anything happened.
This is not a design flaw. It is the reason you can drive a car while holding a conversation, walk into a room and know within a second that something is wrong, or read a client and adjust your approach before you could explain why. Pattern matching at speed is one of the most valuable things the human brain does. It is also the foundation of expertise. Thirty years of experience largely means thirty years of patterns stored so deeply that they run without effort.
Where the patterns come from
Here is the part that matters for the workplace. The brain does not store
patterns as neutral information. It stores them as models of what worked and what was dangerous, together with the emotional and physical state that accompanied them. And it built the earliest and most durable of those models long before anyone had a career.
A child who learned that a raised voice precedes something bad stores that. A child who learned that being questioned means being replaced stores that. A child who learned that visible mistakes cost affection stores that. Decades later, none of this is remembered as an event. It survives as a fast, wordless expectation.
Current models of brain function describe perception less as reception and more as prediction. The brain generates an expectation of what is happening, based on what it has stored, and then uses incoming sensory data mainly to correct that expectation.
You do not first see the situation and then interpret it. You largely see your interpretation.
So when a manager asks for a status update, two different nervous systems can produce two entirely different realities from the same sentence. One receives a request for information. The other receives evidence of being monitored, and the body has already responded before the words are fully processed.
Why stress makes this worse, not better
Under acute stress, the balance shifts. Noradrenaline and cortisol reduce the influence of the prefrontal cortex, the region responsible for deliberate evaluation, perspective taking, and holding several interpretations at once. At the same time, control moves toward faster, habit-based and threat-oriented systems. This is well documented and it is not a malfunction. In genuine danger, you want the fast system to win.
The consequence in an office is that the more pressure someone is under, the more their behaviour is generated by old models and the less it is checked against present evidence.
Stress does not make people irrational. It makes them historical.
Add cognitive load to that. When attention is saturated by parallel projects, notifications, and unfinished decisions, there is less capacity left to notice a prediction error, question it, and update the model. The brain stays with its priors because updating is expensive and nothing is available to pay for it.
The gap
This is the structural problem of the current working world, and it is worth stating plainly.
The human nervous system was shaped over hundreds of thousands of years in environments where context changed slowly. A pattern learned at twenty was still broadly accurate at forty. The value of automatic pattern matching depended on that stability.
Our working environments now change faster than that. Roles, tools, team compositions, communication channels, and expectations shift within months. The pattern that made someone successful in their first leadership position may be actively counterproductive in their current one. Nothing in the biology signals this. Stored programmes do not come with expiry dates. They simply keep running, in a context that no longer matches the one they were built for.
That distance between what the environment demands and what a human body can actually process is the gap. Most of what gets labelled as a personality conflict, a communication problem, or a culture issue is happening inside it.
How projection is produced
Now the mechanism becomes visible. When an old model fires, the emotional charge is real and present, but its source is not. The brain requires a cause for a state this strong, and the most available candidate is the person standing in front of you. So the issue is attributed to the trigger.
The colleague who asked the question becomes the rival from an earlier job. The manager who set the deadline becomes a controlling figure from childhood. From that moment, the other person is no longer perceived as themselves. They have become a projection surface, and every subsequent word, decision, and silence gets filtered through a model they never had access to.
This is why such conflicts survive mediation, new agreements, and clearer processes. Those interventions operate on the factual level, and the conflict is not being fought on the factual level.
What to ask before the next difficult conversation
-
Is the intensity of my reaction proportionate to what actually happened in this room?
-
Do I recognise this feeling from other contexts, including ones that existed long before this working relationship?
-
Am I responding to what this person said and did, or to what they represent to me?
-
What in my reaction belongs to me, without giving up legitimate criticism of real behaviour?
-
And in the other direction: am I currently being perceived as myself, or have I become someone’s projection surface?
None of this dissolves in a single conversation. The stored model does not disappear because it has been recognised. But recognition changes what happens next, because a pattern that has been named can be interrupted, and one that has not been named simply runs.
As long as projections stay unexamined, two people in a meeting are not negotiating with each other. Each is negotiating with a version of the past that the other cannot see.