HCR-Reasoner: Synergizing Large Language Models and Theory for Human-like Causal Reasoning

TL;DR

HCR-Reasoner presents a unified framework that combines actual causality formalisms with causal-judgment theory to simulate human-like causal reasoning in large language models. The approach uses a three-stage process: establish a causal setting, infer factor values (actual/necessary/sufficient causes, normality, and intention), and apply theory-guided algorithmic reasoning to produce a Yes/No judgment plus an explanation. A dedicated benchmark, HCR-Bench, with 1,093 annotated instances and detailed reasoning steps, enables fine-grained evaluation beyond existing causal datasets. Experimental results show that grounding LLM reasoning in explicit theory yields consistent, significant improvements across diverse models, often surpassing human averages on causal-judgment tasks. The work demonstrates that explicit theory integration can yield faithful, human-like causal reasoning in AI systems, with practical implications for interpretable AI and safer decision-making.

Abstract

Genuine human-like causal reasoning is fundamental for strong artificial intelligence. Humans typically identify whether an event is part of the causal chain first, and then influenced by modulatory factors such as morality, normality, and intention to make the final judgment. These two stages naturally map to the fields of 1) actual causality that provides formalisms for causal chain membership and 2) causal judgment from cognitive science that studies psychological modulators that influence causal selection. However, these two domains have largely been studied in isolation, leaving a gap for a systematic method based on LLMs. Therefore, we introduce HCR-Reasoner, a framework that systematically integrates the theory of actual causality and causal judgment into LLMs for human-like causal reasoning. It simulates humans by using actual causality formalisms to filter for structurally necessary candidate causes and causal judgment factors to determine the psychologically selected cause. For fine-grained evaluation, we introduce HCR-Bench, a challenging benchmark with 1,093 annotated instances with detailed reasoning steps. Results show HCR-Reasoner consistently and significantly improves LLMs' causal alignment with humans, and that explicitly integrating theory-guided reasoning into LLMs is highly effective for achieving faithful human-like causal reasoning.

Figures (5)

• Figure 1: An example in HCR-Bench. To simulate human-like causal reasoning, one needs to consider both actual causality formalisms and causal judgment factors.
• Figure 2: The overview of HCR-Reasoner. The causal structure at the bottom is not actually constructed.
• Figure 3: Fine-grained accuracies of different factors.
• Figure 4: Results of the causal analysis.
• Figure 5: The causal representation of the preemption case.

Theorems & Definitions (12)

• Definition 1
• Definition 2
• Definition 3
• Proposition 1
• proof
• proof
• proof
• proof
• proof
• proof