---
title: "Cognitive Bias Signatures in Search-Equipped LLM Agents: A Stage-Resolved Analysis"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2611.05720v1"
permalink: "https://feed.astralor.com/items/itm_97b0ec4c4f9799727d96cc23"
published: "2026-11-03T22:46:55.000Z"
score: 57
channel: "ai"
tags: ["LLM","AI Agent","RAG","Cognitive Bias Signatures","Search-Equipped LLM Agents","Stage-Resolved Analysis Search","Drawing","Search"]
---

# Cognitive Bias Signatures in Search-Equipped LLM Agents: A Stage-Resolved Analysis

## At a glance

Search can help LLM agents correct mistakes, but their starting conditions may shape which evidence they find and use. Drawing on cognitive-bias paradigms, we develop a controlled search harness to examine result order,…

## Key points

- Search can help LLM agents correct mistakes, but their starting conditions may shape which evidence they find and use.
- Drawing on cognitive-bias paradigms, we develop a controlled search harness to examine result order, records of prior work, and initial hypotheses while holding each question and document collection fixed.
- Matched controls distinguish order effects from differences in document composition and the effects of failed work from context length.

## Full text

Search can help LLM agents correct mistakes, but their starting conditions may shape which evidence they find and use. Drawing on cognitive-bias paradigms, we develop a controlled search harness to examine result order, records of prior work, and initial hypotheses while holding each question and document collection fixed. Matched controls distinguish order effects from differences in document composition and the effects of failed work from context length. Across 10 models and 1,000 synthetic search worlds, we track queries, retrieved documents, document openings, and final answers. Our comparisons suggest that final answers are more sensitive to initial evidence order during active search than when all evidence is supplied in full. Dead-end histories can reduce strategy switching relative to matched neutral histories, indicating an effect beyond context length. Under wrong hypotheses, stage measurements distinguish reduced retrieval or opening from errors that persist after counter-evidence is opened. Exploratory interventions show that opening counter-evidence more often can coincide with lower accuracy, while accuracy can improve with little change in opening.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_97b0ec4c4f9799727d96cc23