Analyzing Cognitive Differences Among Large Language Models through the Lens of Social Worldview

TL;DR

This paper introduces the Social Worldview Taxonomy to analyze and quantify the socio-cognitive attitudes of large language models, revealing their cognitive profiles and adaptability to social cues, which has implications for AI transparency and responsibility.

Contribution

It presents a novel evaluation framework based on Cultural Theory to assess the implicit social worldviews of LLMs and demonstrates their cognitive flexibility through social cue modulation.

Findings

Distinct cognitive profiles identified across 28 LLMs

Explicit social cues influence model responses systematically

Models exhibit latent cognitive flexibility in social contexts

Abstract

Large Language Models significantly influence social interactions, decision-making, and information dissemination, underscoring the need to understand the implicit socio-cognitive attitudes, referred to as "worldviews", encoded within these systems. Unlike previous studies predominantly addressing demographic and ethical biases as fixed attributes, our study explores deeper cognitive orientations toward authority, equality, autonomy, and fate, emphasizing their adaptability in dynamic social contexts. We introduce the Social Worldview Taxonomy (SWT), an evaluation framework grounded in Cultural Theory, operationalizing four canonical worldviews, namely Hierarchy, Egalitarianism, Individualism, and Fatalism, into quantifiable sub-dimensions. Through extensive analysis of 28 diverse LLMs, we identify distinct cognitive profiles reflecting intrinsic model-specific socio-cognitive structures. Leveraging principles from Social Referencing Theory, our experiments demonstrate that explicit social cues systematically modulate these profiles, revealing robust patterns of cognitive adaptability. Our findings provide insights into the latent cognitive flexibility of LLMs and offer computational scientists practical pathways toward developing more transparent, interpretable, and socially responsible AI systems