---
title: "halo effect"
type: "entity"
entity_kind: "concept"
status: "hub"
canonical_name: "halo effect"
aliases: ["halo error","constant error (Thorndike)"]
first_seen: "2026-08-03T00:00:00.000Z"
writer_model: "claude-opus-4-8"
connects_to: ["cognitive bias","Edward L. Thorndike","source-reliability vs credibility independence","intelligence tradecraft (source evaluation)","reliability-aware RAG"]
seek_code_commit: "46c6736"
---


The halo effect is the tendency for a single global impression of a person or thing to bleed into ratings of its specific, supposedly independent attributes, so that trait judgments correlate far more tightly than genuine independence would produce. The term *halo* was coined by [[entity-edward-thorndike|Edward L. Thorndike]] in his 1920 paper "A Constant Error in Psychological Ratings," from WWI officers' ratings of aviation cadets; the fixed compound *halo effect* in common use today is a later crystallization, not Thorndike's printed wording (his paper says only "the constant error of the 'halo'"). It matters to this vault as the named mechanism sitting under an entire cluster of modern source-evaluation notes that describe the same failure without invoking it.

## References
- [[claim-thorndike-1920-halo-effect-ratings-too-high-and-too-even]] — the origin finding and coinage
- [[observation-halo-effect-names-the-2025-reliability-credibility-leak]] — the 105-year bridge to 2025 work
- [[claim-source-reliability-and-credibility-are-not-judged-independently]] — the modern intelligence-analysis instance
- [[claim-reliability-aware-rag-estimates-source-reliability-separately-from-relevance]] — the RAG instance
- [[observation-intelligence-doctrine-and-rag-independently-derived-a-two-axis-source-model]] — the two-axis convergence the halo effect underlies
- Capture: 10-inbox/raw/2026-08-03-hop-thorndike-halo-effect.md