This is a Lexicon entry: the mechanism, when it shows up, and the countermeasure. The full correction, with real-world cases and the audit prompt, hasn't been written yet.
Definition
The challenge of distinguishing meaningful information (signal) from irrelevant random variation (noise) in any dataset, model or stream of observations. The signal-to-noise ratio is a fundamental constraint on learning from data: too much noise makes the signal undetectable; overfitting to noise produces models that work on historical data and fail on new data. In everyday reasoning, the same problem appears as the clustering illusion - seeing patterns in what is actually noise.
When it shows up
In any dataset or information stream; when distinguishing genuine patterns from random variation. In any expert field: when understanding your own calibration helps you communicate uncertainty to stakeholders. In research: when hypothesis pre-registration forces you to articulate your prior before seeing data.
Failure mode
Seeing signal in noise and acting on it. Building theories from statistical artefacts. Responding to random variation as if it were meaningful.
Countermeasure
Ask: "What portion of this variation is likely random?" Require replication and mechanism before treating any pattern as real signal.
Related traps
Also connected in the map3 more, locked
Credit & first seen
DiscoveredA foundational concept in signal processing (Shannon, 1948) and statistics; discussed in forecasting by Nate Silver in The Signal and the Noise (2012). source ↗
Also revealed
Entry #559 of 631 in The Lexicon · see the full Lexicon