Do LLM Agents Mirror Socio-Cognitive Effects in Power-Asymmetric Conversations?

TL;DR

This study investigates whether large language models exhibit socio-cognitive effects of power, such as language coordination and authority bias, in simulated power-asymmetric dialogues, revealing both similarities and nuances compared to human behaviors.

Contribution

It demonstrates that LLMs can mirror key socio-cognitive effects of power in dialogues, highlighting both their capabilities and potential safety concerns.

Findings

LLMs exhibit language coordination and pronoun usage patterns linked to power status.

Power effects influence persuasion success and compliance with unsafe requests.

Results reveal variability and nuances in LLMs' socio-cognitive behaviors.

Abstract

Power differences shape human communication through well documented socio cognitive effects, including language coordination, pronoun usage, authority bias, and harmful compliance. We examine whether large language models (LLMs) exhibit similar behaviors when assigned high or low status personas. Using personas from diverse professions, we simulate multi turn, power asymmetric dialogues (e.g., principal teacher, justice lawyer) and measure (i) language coordination, (ii) pronoun usage, (iii) persuasion success, and (iv) compliance with unsafe requests. Our results show that LLMs show key socio-cognitive effects of power, albeit with nuances and variability, linking simulated interactions to both desirable and unsafe behaviors.