Evaluating LLM Alignment With Human Trust Models

TL;DR

This paper investigates how large language models internally conceptualize trust by analyzing their activation space, revealing alignment with human socio-cognitive trust models, which informs AI-human collaboration.

Contribution

It introduces a white-box method to analyze LLM trust representations and identifies their alignment with specific human trust models, advancing understanding of AI social cognition.

Findings

EleutherAI/gpt-j-6B aligns most with Castelfranchi's socio-cognitive trust model.

The model's trust representation also shows significant similarity to the Marsh Model.

Results support the idea that LLMs encode social cognition constructs.

Abstract

Trust plays a pivotal role in enabling effective cooperation, reducing uncertainty, and guiding decision-making in both human interactions and multi-agent systems. Although it is significant, there is limited understanding of how large language models (LLMs) internally conceptualize and reason about trust. This work presents a white-box analysis of trust representation in EleutherAI/gpt-j-6B, using contrastive prompting to generate embedding vectors within the activation space of the LLM for diadic trust and related interpersonal relationship attributes. We first identified trust-related concepts from five established human trust models. We then determined a threshold for significant conceptual alignment by computing pairwise cosine similarities across 60 general emotional concepts. Then we measured the cosine similarities between the LLM's internal representation of trust and the derived trust-related concepts. Our results show that the internal trust representation of EleutherAI/gpt-j-6B aligns most closely with the Castelfranchi socio-cognitive model, followed by the Marsh Model. These findings indicate that LLMs encode socio-cognitive constructs in their activation space in ways that support meaningful comparative analyses, inform theories of social cognition, and support the design of human-AI collaborative systems.