A Survey on LLMs from a Human-Centric Perspective

The rapid evolution of large language models (LLMs) has positioned them as powerful computational instruments for human-centered research and applications, spanning tasks involving cognition, decision-making, and social interaction. This survey examines LLMs from a human-centric perspective organized around individual and collective levels. We first review LLM competencies across perception, analysis, integration, execution, and social interaction, highlighting both their demonstrated capabilities and the evaluation conditions under which these capabilities are observed. We then explore how LLMs are used across behavioral science, psychology, linguistics, political science, economics, and sociology as research tools, behavioral simulators, and objects of inquiry, covering both individual behavior and collective social processes. Finally, we identify challenges and future research directions, such as improving LLM adaptability, affective-to-behavior evaluation, and cultural sensitivity, while addressing inherent biases and enhancing frameworks for human-AI collaboration. This survey aims to provide a foundational understanding of LLMs from a human-centric perspective, offering insights into their current capabilities and potential for future development.