---
id: "prereq-bayesian-agent-theory"
type: "prereq"
source_timestamps: ["¶8"]
tags: ["statistics", "economics"]
related: ["quote-bayesian-agents", "concept-willful-ignorance-in-ai"]
reason: "Necessary to understand Chan's quote contrasting theoretical human rationality with actual strategic, motivated ignorance."
sources: ["adoption"]
sourceVaultSlug: "hbr-seg-adoption"
originDay: 9
articleStem: "hbr-edu-37-employees-not-questioning-ai"
sourceUrl: "https://hbr.org/2026/06/employees-arent-questioning-ai-advice-enough"
sourceTitle: "Employees Aren’t Questioning AI Advice Enough"
---
# Bayesian Agent Theory

**Why you need this:** Necessary to understand [[quote-bayesian-agents|Chan's quote]] contrasting theoretical human rationality with actual strategic, motivated ignorance.

The source assumes the reader understands what a **'perfectly rational Bayesian agent'** is — a theoretical entity in economics and statistics that updates its beliefs *optimally* based on all available new evidence. Understanding this highlights the contrast with actual human behavior, which selectively ignores evidence, as captured in [[concept-willful-ignorance-in-ai]].

**Enrichment / adjacent literature:** The contrast is grounded in standard Bayesian decision theory. The empirical literature documents systematic deviations from optimal evidence acquisition — status quo bias, selective exposure, and motivated reasoning — which is exactly the gap Chan's experiment exploits when it shows people undervalue explanations even when those explanations would improve accuracy.