---
name: Perceived Threat
slug: perceived-threat
type: cognitive-emotional process
status: running
version: 4.2.1
released: "~200,000 BCE"
maintainer: amygdala
dependencies:
  - fear
  - pattern-recognition
  - memory
  - context (optional, frequently missing)
license: Inherited. Cannot be uninstalled.
tags:
  - survival
  - cognition
  - anxiety
  - false-positives
  - threat-detection
---

Perceived Threat

What it actually is

A signal fired when something might be dangerous, evaluated by a system optimized for the Pleistocene and running, unchanged, in the present.

How it works

The process initiates before conscious thought arrives. The amygdala flags a stimulus. The body responds. Cortisol ships. Heart rate adjusts. Only then does the prefrontal cortex receive the ticket and begin asking reasonable questions like "was that actually a threat, or just a colleague's tone in a Slack message?"

By then, the damage is already logged.

The pipeline:

Stimulus detected (real, imagined, or misread)
Pattern matched against memory archive, including outdated entries
Threat level assigned (0 to critical, skewed toward critical)
Physiological response dispatched
Narrative constructed retroactively to justify response
Rational review, if invoked at all, arrives late

Step 5 is where most of the interesting problems live.

Features

Speed: Threat assessment completes in approximately 12 milliseconds. Accuracy not guaranteed.
Persistence: Signal can remain active long after the stimulus has resolved, changed, or turned out to be a coat on a chair.
Generalization: One bad instance can flag an entire category. One dog, one decade.
Social extension: Perceived threat includes threats to status, belonging, identity, and the sense that someone is quietly judging you.

Known bugs

False positive rate: Extremely high in low-stakes modern environments. The system cannot distinguish a predator from a performance review.
Context blindness: The context dependency is listed as optional and is frequently not loaded.
Threat inflation: Under chronic stress, baseline sensitivity increases. Everything starts to feel like the thing before the thing that kills you.
Mirror misread: Perceived threats from others sometimes originate as unresolved internal states. The bug reports itself as external.

"I wasn't scared. I was just preparing for something that statistically could have happened." — every user, always

Error codes

E001  THREAT_UNVERIFIED       Stimulus flagged without sufficient evidence
E002  CONTEXT_NOT_LOADED      Response fired without environmental data
E003  MEMORY_OUTDATED         Pattern matched to archived entry, no longer valid
E004  SOCIAL_THREAT_OVERFLOW  Status anxiety exceeding available working memory
E005  LOOP_DETECTED           Perceived threat generating new perceived threats

Edge cases

The threat was real and the response was correct. This happens. The system exists for a reason.
The perceived threat belongs entirely to the past but is being rendered in the present at full resolution.
Two people in the same room both register each other as threats simultaneously, with no external cause. Known as conflict.

FAQ

Can it be turned off? No. Volume can be adjusted through therapy, somatic work, or extended safety experiences. The process itself is load-bearing infrastructure.

Is it rational? It is functional. Rationality is a different module with slower load times.

What if the threat is real? Then this process worked exactly as intended. Update the changelog accordingly.

drafting spec…