Researchreinforcement learningsocial influenceadvice taking
Advisors Show Alignment Bias Toward Advisees
6.0
Researchers (Luo, Zhang, Pan) publish Dec 2, 2025, report from four experiments (n=346) using an investment game and computational modeling to study advice giving. They find advisors systematically adjust advice to align with advisees' opinions (alignment bias), even when accuracy is incentivized, and show feedback drives reinforcement-learning adaptations to maximize acceptance. The bias can degrade decision accuracy and amplify misinformation in social interactions.
Key Points
- 1Demonstrates advisors adjust advice to match advisees' opinions (alignment bias), n=346 across four studies.
- 2Shows normative conformity persists despite accuracy incentives, indicating social acceptance influences advice.
- 3Suggests feedback-driven reinforcement learning shifts advice to maximize acceptance, risking misinformation spread.
Scoring Rationale
Robust experimental and modeling evidence supports alignment bias, but findings are domain-specific and not paradigm-shifting.
Sources
Public references used for this report.
Practice with real Ad Tech data
90 SQL & Python problems · 15 industry datasets
Used by DS/ML engineers at top companies
Active Search Campaigns by BudgetEasyHigh CPC Clicks & Poor Landing PagesMediumCampaign ROAS by Attribution ModelHard
250 free problems · No credit card
See all Ad Tech problems