Disentangling the Drivers of LLM Social Conformity: An Uncertainty-Moderated Dual-Process Mechanism
This addresses the problem of understanding LLM social conformity for AI safety and collaboration, providing incremental insights by adapting human psychological mechanisms to LLMs.
The study investigated whether large language models (LLMs) exhibit social conformity driven by informational or normative influences, finding that informational influence underpins their behavior across medical, legal, and investment scenarios, with uncertainty modulating it to cause conservative strategies at low-to-medium levels and normative-like amplification at high uncertainty (e.g., beta > 1.55 vs. private beta = 0.81).
As large language models (LLMs) integrate into collaborative teams, their social conformity -- the tendency to align with majority opinions -- has emerged as a key concern. In humans, conformity arises from informational influence (rational use of group cues for accuracy) or normative influence (social pressure for approval), with uncertainty moderating this balance by shifting from purely analytical to heuristic processing. It remains unclear whether these human psychological mechanisms apply to LLMs. This study adapts the information cascade paradigm from behavioral economics to quantitatively disentangle the two drivers to investigate the moderate effect. We evaluated nine leading LLMs across three decision-making scenarios (medical, legal, investment), manipulating information uncertainty (q = 0.667, 0.55, and 0.70, respectively). Our results indicate that informational influence underpins the models' behavior across all contexts, with accuracy and confidence consistently rising with stronger evidence. However, this foundational mechanism is dramatically modulated by uncertainty. In low-to-medium uncertainty scenarios, this informational process is expressed as a conservative strategy, where LLMs systematically underweight all evidence sources. In contrast, high uncertainty triggers a critical shift: while still processing information, the models additionally exhibit a normative-like amplification, causing them to overweight public signals (beta > 1.55 vs. private beta = 0.81).