← heapsort-ai

cognitive bias

5 items

ARTICLEDEV.to AI·4/17/2026

Projection 2.0: How We Attribute Personality, Gender, and Intent to Models Based on Tiny Prompt Variations

"Projection 2.0" describes the human tendency to attribute personality, gender, and intent to AI systems based on tiny variations in how they are addressed. This article explores this fascinating quirk of human psychology, its implications for AI design and ethics, and the importance of becoming more conscious of our own projections.

27
RESEARCHarXiv CS.CL·4/15/2026

Narrative over Numbers: The Identifiable Victim Effect and its Amplification Under Alignment and Reasoning in Large Language Models

This research systematically investigates the Identifiable Victim Effect (IVE) in Large Language Models, a cognitive bias where specific, narratively described victims receive more resources than statistically characterized groups. The large-scale empirical study across 16 frontier LLMs determines if these systems inherit human affective irrationalities in critical applications like humanitarian triage and content moderation.

27
ARTICLEDEV.to AI·28d ago

赛仑

This article uses the "Siren" metaphor to analyze how modern algorithms and the attention economy exploit cognitive biases to create "attention black holes". It provides strategies for individuals to resist these digital traps, such as cognitive decentralization and information minimalism.

27
RESEARCHarXiv CS.CL·4/6/2026

Failing to Falsify: Evaluating and Mitigating Confirmation Bias in Language Models

Este estudo investiga o viés de confirmação em grandes modelos de linguagem (LLMs) usando uma tarefa de descoberta de regras, revelando que os LLMs exibem essa tendência, o que retarda a descoberta de regras ocultas. Ele demonstra que estratégias de intervenção, como o uso de prompts específicos, podem consistentemente diminuir esse viés.

27