Autopsy of the Absent
Error LOL : Pattern not Found
Sat, 9 Aug 2025
Today, I aim the scalpel at my own favourite blades: intuition and pattern recognition.
Disclaimer: The dissection may not bleed clarity, but some acts of violence are too elegant to resist.
Even a necessary wound deserves a sterile edge. So first, the courtesy of acknowledgment: intuitive pattern recognition is far from useless, it’s a habit that has written our survival story. Human thought orbits around it, spotting repetition, tying them to past experience, and categorising the unfamiliar into the familiar.
With enough repetition, categories congeal into patterns, patterns ossify into narratives and narratives loop back to reinforce the very patterns that they arose from. They are like cognitive handrails, providing an illusion of stability in ceaseless flux. Static mental models that reduce the cost of storing, retrieving, and acting on information. The brain favours stable representations because tracking high variance inputs consumes far more working memory and attention.
It has also served us well in the evolutionary game: repeated random events, like a rustle in the bushes, became symbols when organisms learn to associate meaning with them, here, “predator” because the cost of a false negative is far higher than a false positive "wind". False positives are, quite literally, why I’m here to write this.
So here's the first cut, for all its utility, human pattern recognition is at its core a data compression system. A tidy exercise in fabricating sequential logic, put together from sparse cues collected from prior experience, good enough for rapid decision making under a certain uncertainty, but dissolves when more information floods in.
But a sudden climax like this without any buildup is indecent, violation even, so we’ll go about this slowly.
Now let me take you where the tissue is most thin, where seeing one pattern pierces another's skin. Our pattern-seeking mind can sometimes spin self-reinforcing loops, when what we notice or imagine shapes our behavior just enough to make the pattern come true.
We know the script of a self-fulfilling prophecy. You see a teammate away from their desk more than usual and decide, “They’re not serious about work” Maybe they’re just efficient and don’t need as much desk time, but your “recognised pattern” is set. So you stop delegating, start hoarding work, and micromanage whatever you do hand off. Predictably, they disengage, motivation drops, and deadlines slip. Then you point to the misses as proof your instinct was right. But that wasn’t foresight, it was the micromanaging.
Our hunger for patterns breeds apophenia, seeing patterns when there are none, and its close ally, confirmation bias. Spot a pattern, and you cherry-pick evidence to keep it alive. Astrology I’m looking at you. Once a hypothesis takes root, every matching detail feels like proof, each omission quietly erased. Pair that with the mechanics of a self-fulfilling prophecy, and its alleged “accuracy” often comes from our own biased interventions, editing reality to fit the script, not from any cosmic foresight. The result is a closed, self-reinforcing loop as legitimate as any other system humans have devised.
Sometimes they're just harmless illusions, pareidolia in clouds or 11:11 on the clock. Every wish revives the thought, anchoring it deeper in your headspace, priming your actions to align. And yes, 11:11 works, obviously. Synchronicity, and all that.
Randomness is cognitively expensive, and the moment something repeats, it triggers an instinct to stabilise it with meaning. This compulsion once sharpened our chances of survival, now it underwrites almost everything we do. But as apophenia reminds us, not everything means something.
Now that the superficials are peeled back, let’s hunt what hides in the marrow. Patterns aren’t universal laws, think of them as a local contract, an agreement between all the participants inside a given system. Its coherence comes from internal reciprocity, each idea or symbol is validated by the others it connects to, upheld by the system’s shared framework and inherited memory. Within that frame, every link reinforces the next in a self-sustaining loop. Step outside, into another culture, discipline, or worldview and the same pattern may seem arbitrary, irrelevant, or nonsensical, because its terms were never negotiated there.
Language systems offer an easy example, words hold meaning only because speakers collectively agree they do. Outside that shared system, same pattern becomes noise.
Cultures filter raw data into wildly different constellations, literally. The same stars form different shapes across civilizations. Move closer to one in space and the illusion collapses, the pattern existed only from a particular vantage point on Earth and you find there is no such thing as a constellation.
And this gets interesting here, we all are working from the same raw data, yet our patterns keep diverging because different local contracts offer different frameworks. Pattern recognition is never a neutral mirror; it’s designed to fit the frameworks we already hold, for maximum coherence. And those frameworks are assembled from only fragments rather than the entire spectrum of available information.
In 1600, astronomers could convincingly explain planets traced perfect circles on invisible crystal spheres around Earth. By 1700, they moved in ellipses around the Sun through empty space. Same sky, different story. As frameworks shift and expand, today’s certainty will be tomorrow’s joke.
And I’m not here to discredit them all, just to deconstruct the machinery. These contracts aren’t “false” so much as conditional. A cardiologist can predict a heart attack because the signs fit within a collective framework of biological knowledge, a system rooted in the body’s own pattern memory at the cellular level. Fortunately not all loops are born in the mind.
Take the spiral in a sunflower’s seed head, the swirl of a galaxy, the recursive elegance of fractals. They feel immutable, until you realize, they too are only local contracts within their systems. Change the system, and they no longer hold. Sunflower on the ISS growing in microgravity abandon their perfect spirals. Snowflakes deform into asymmetrical blobs without stable temperature and humidity. Dams can distort river fractals. Galaxy collisions twist spiral arms into chaotic smears. (Hello Centaurus A.)
No pattern is invincible, only as persistent as the system that sustains it. Call their breaking a pattern too, and we can spar until we reach the edge of epistemic recursion, if there's ever such a thing. But for now, stay with me.
Which brings us to an interesting question, what if there are other realities, where our logic is useless, and every pattern we’ve built to define it breaks?
We don’t need a parallel universe to watch this play out; AI gives us a near-perfect simulation of this. Large language models are pure pattern engines. Feed them data, they predict the next word, the next label. In one well known experiment, an AI trained to tell wolves from huskies learned not animals but snow: wolf photos had snowy backgrounds, husky photos didn’t. Snow became its shortcut. Show it a picture of an empty snowy field, and it would still confidently “see” a wolf.
It’s a neat reminder of how pattern-recognition systems can latch onto the wrong signal, one that works perfectly within their narrow training data but fails outside it.
Humans aren’t that different. Our “truths” emerge from our own incomplete training data. Let me move to a more agnostic stance and then there are patterns that simply exist beyond our perception, some might even contradict the patterns we swear by. Think of the shift from the Ptolemaic to the Newtonian model of planetary motion we touched on earlier.
Generative AI holds a mirror to the fractures running through our pattern-seeking bones. LLMs also don’t understand truth, they generate the most probable sequences of words based on patterns in their training data. When an answer feels “validated,” it’s often because it traces recurrence, not because any fact was verified. Ask the same open-ended question twice and you may get entirely different, yet equally coherent, answers. Each is a plausible narrative assembled on the fly, leaving us to wonder, which, if any, deserves belief?
And doesn't reality work the same way? Every experience arrives raw, unlabelled, from it, countless interpretations are possible. We choose whatever feels familiar in our ongoing narrative loop, and call it truth. But truth isn’t singular, only our chosen pattern is.
This is where you might ask: so what do we do when there’s no “right” pattern at all? Enter cognitive dissonance, my favourite playground. What happens when multiple narrative frameworks coexist, and they all seem to make sense? Resolution usually comes when one narrative offers a cleaner, more coherent sequence. But as we’ve already seen, coherence doesn’t guarantee truth, it’s just the best-fitting story within a particular system of thought.
And yes, this very exploration sits inside the same, scanning randomness for a repeating evidence of absence, inventing as much meaning as it wants to discover. If you happened to question it, good. If you didn’t , also good. You’ve simply spared yourself the spiral of doubting it. Either way, we end where we began. Neat loop.
So the point? None, except the one you can convince yourself of. But I warned you, there will be no closure. I was bored, this was fun, now I’m tired. And since Shakespeare didn’t live long enough to know ghosting, these violent delights will just have to settle for silent ends.