Do Large Language Models Reason Causally Like Us? Even Better?
On the map: LLM reasoning → Causal reasoning in LLMs
Abstract (arXiv)
Causal reasoning is a core component of intelligence. Large language models (LLMs) have shown impressive capabilities in generating human-like text, raising questions about whether their responses reflect true understanding or statistical patterns. We compared causal reasoning in humans and four LLMs using tasks based on collider graphs, rating the likelihood of a query variable occurring given evidence from other variables. LLMs' causal inferences ranged from often nonsensical (GPT-3.5) to human-like to often more normatively aligned than those of humans (GPT-4o, Gemini-Pro, and Claude). Computational model fitting showed that one reason for GPT-4o, Gemini-Pro, and Claude's superior performance is they didn't exhibit the "associative bias" that plagues human causal reasoning. Nevertheless, even these LLMs did not fully capture subtler reasoning patterns associated with collider graphs, such as "explaining away".
Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Placed in Causal reasoning in LLMs, one of 17 directions in LLM reasoning. Builds on 9 of its 29 references that are in the map. First posted on arXiv 14 Feb 2025.
Builds on
9 of 29 references.
Closest ancestors · 2
Same direction, newest first
- 2023CLadder: Assessing Causal Reasoning in Language Models7 Dec
- 2023Causal Reasoning and Large Language Models: Opening a New Frontier for Causality28 Apr
Other references · 7
- 2023Sparks of Artificial General Intelligence: Early experiments with GPT-4cited by 2,579
- 2023Capabilities of GPT-4 on Medical Challenge Problemscited by 630
- 2024GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Modelscited by 533
- 2023Causal Parrots: Large Language Models May Talk Causality But Are Not Causalcited by 146
- 2024A Peek into Token Bias: Large Language Models Are Not Yet Genuine Reasonerscited by 73
- 2024Apple Intelligence Foundation Language Modelscited by 22
- 2023Borges and AIcited by 3
Built on by
Citing papers per year: 2025: 3; 2026: 1.
4 papers.
- 2025HCR-Reasoner: Synergizing Large Language Models and Theory for Human-like Causal Reasoningcited by 1
- 2025Causal Process Models: Reframing Dynamic Causal Graph Discovery as a Reinforcement Learning Problemcited by 1
- 2026Position: Behavioral Systems Require Behavioral Testscited by 1
- 2025If Probable, Then Acceptable? Understanding Conditional Acceptability Judgments in Large Language Modelscited by 0
- 2026Position: Behavioral Systems Require Behavioral Tests31 May 2026
- 2025If Probable, Then Acceptable? Understanding Conditional Acceptability Judgments in Large Language Models9 Oct 2025
- 2025Causal Process Models: Reframing Dynamic Causal Graph Discovery as a Reinforcement Learning Problem18 Jul 2025
- 2025HCR-Reasoner: Synergizing Large Language Models and Theory for Human-like Causal Reasoning13 May 2025
Similar, not cited
Close in wording, no citation either way
- 2026Do LLMs Share Human-Like Biases? Causal Reasoning Under Prior Knowledge, Irrelevant Context, and Varying Compute Budgets
- 2025Unveiling Causal Reasoning in Large Language Models: Reality or Mirage?
- 2023Do large language models and humans have similar behaviors in causal inference with script knowledge?
- 2025Causal Strengths and Leaky Beliefs: Interpreting LLM Reasoning via Noisy-OR Causal Bayes Nets
- 2025Causality for Natural Language Processing
- 2024Large Language Models and Causal Inference in Collaboration: A Survey
- 2024How Likely Do LLMs with CoT Mimic Human Reasoning?
- 2023Can Large Language Models Infer Causation from Correlation?