# Using Counterfactual Tasks to Evaluate the Generality of Analogical Reasoning in Large Language Models

arXiv: https://arxiv.org/abs/2402.08955 · HTML: https://mlsanity.com/paper/2402.08955/

## Key facts

- Cited by 38 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning.
- Builds on 8 of its 16 references that are in the map.
- First posted on arXiv 14 Feb 2024.

## Placed in

- LLM reasoning / [Reasoning benchmarks and evaluation](https://mlsanity.com/lr/reasoning-benchmarks-and-evaluation/)

## Builds on (most cited first, 8 of 8)

- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Emergent Abilities of Large Language Models](https://mlsanity.com/paper/2206.07682/) (2022)
- [Towards Reasoning in Large Language Models: A Survey](https://mlsanity.com/paper/2212.10403/) (2022)
- [Faith and Fate: Limits of Transformers on Compositionality](https://mlsanity.com/paper/2305.18654/) (2023)
- [Reasoning or Reciting? Exploring the Capabilities and Limitations of Language Models Through Counterfactual Tasks](https://mlsanity.com/paper/2307.02477/) (2023)
- [Emergent Analogical Reasoning in Large Language Models](https://mlsanity.com/paper/2212.09196/) (2022)
- [Embers of Autoregression: Understanding Large Language Models Through the Problem They are Trained to Solve](https://mlsanity.com/paper/2309.13638/) (2023)
- [Response: Emergent analogical reasoning in large language models](https://mlsanity.com/paper/2308.16118/) (2023)

## Built on by (most cited first, 30 of 38)

- [Alice in Wonderland: Simple Tasks Showing Complete Reasoning Breakdown in State-Of-the-Art Large Language Models](https://mlsanity.com/paper/2406.02061/) (2024)
- [Not All LLM Reasoners Are Created Equal](https://mlsanity.com/paper/2410.01748/) (2024)
- [LLM-Driven Robots Risk Enacting Discrimination, Violence, and Unlawful Actions](https://mlsanity.com/paper/2406.08824/) (2024)
- [Achieving Time Series Reasoning Requires Rethinking Model Design, Tasks Formulation, and Evaluation](https://mlsanity.com/paper/2502.01477/) (2025)
- [LLMs as Models for Analogical Reasoning](https://mlsanity.com/paper/2406.13803/) (2024)
- [A Systematic Assessment of OpenAI o1-Preview for Higher Order Thinking in Education](https://mlsanity.com/paper/2410.21287/) (2024)
- [Evidence from counterfactual tasks supports emergent analogical reasoning in large language models](https://mlsanity.com/paper/2404.13070/) (2024)
- [Assessing the Reasoning Capabilities of LLMs in the context of Evidence-based Claim Verification](https://mlsanity.com/paper/2402.10735/) (2024)
- [Towards Learning Abductive Reasoning using VSA Distributed Representations](https://mlsanity.com/paper/2406.19121/) (2024)
- [Can Large Language Models generalize analogy solving like children can?](https://mlsanity.com/paper/2411.02348/) (2024)
- [A Review of Multimodal Explainable Artificial Intelligence: Past, Present and Future](https://mlsanity.com/paper/2412.14056/) (2024)
- [From Text to Life: On the Reciprocal Relationship between Artificial Life and Large Language Models](https://mlsanity.com/paper/2407.09502/) (2024)
- [A Survey of Inductive Reasoning for Large Language Models](https://mlsanity.com/paper/2510.10182/) (2025)
- [Revealing emergent human-like conceptual representations from language prediction](https://mlsanity.com/paper/2501.12547/) (2025)
- [None of the Others: a General Technique to Distinguish Reasoning from Memorization in Multiple-Choice LLM Evaluation Benchmarks](https://mlsanity.com/paper/2502.12896/) (2025)
- [Evaluating Compositional Scene Understanding in Multimodal Generative Models](https://mlsanity.com/paper/2503.23125/) (2025)
- [A Neuropsychologically Grounded Evaluation of LLM Cognitive Abilities](https://mlsanity.com/paper/2603.02540/) (2026)
- [Anthropocentric bias in language model evaluation](https://mlsanity.com/paper/2407.03859/) (2024)
- [DeFine: Decision-Making with Analogical Reasoning over Factor Profiles](https://mlsanity.com/paper/2410.01772/) (2024)
- [Flying Pigs, FaR and Beyond: Evaluating LLM Reasoning in Counterfactual Worlds](https://mlsanity.com/paper/2505.22318/) (2025)
- [Modelling Analogies and Analogical Reasoning: Connecting Cognitive Science Theory and NLP Research](https://mlsanity.com/paper/2509.09381/) (2025)
- [Empowerment Gain and Causal Model Construction: Children and adults are sensitive to controllability and variability in their causal interventions](https://mlsanity.com/paper/2512.08230/) (2025)
- [Aligning Generalisation Between Humans and Machines](https://mlsanity.com/paper/2411.15626/) (2024)
- [Visual Reasoning in Object-Centric Deep Neural Networks: A Comparative Cognition Approach](https://mlsanity.com/paper/2402.12675/) (2024)
- [Large Language Models, and LLM-Based Agents, Should Be Used to Enhance the Digital Public Sphere](https://mlsanity.com/paper/2410.12123/) (2024)
- [DeduCE: Deductive Consistency as a Framework to Evaluate LLM Reasoning](https://mlsanity.com/paper/2504.07080/) (2025)
- [Analogical Reasoning Within a Conceptual Hyperspace](https://mlsanity.com/paper/2411.08684/) (2024)
- [LLM world models are mental: Output layer evidence of brittle world model use in LLM mechanical reasoning](https://mlsanity.com/paper/2507.15521/) (2025)
- [AssoCiAm: A Benchmark for Evaluating Association Thinking while Circumventing Ambiguity](https://mlsanity.com/paper/2509.14171/) (2025)
- [DEVAL: A Framework for Evaluating and Improving the Derivation Capability of Large Language Models](https://mlsanity.com/paper/2511.14813/) (2025)

## Abstract (arXiv)

Large language models (LLMs) have performed well on several reasoning benchmarks, including ones that test analogical reasoning abilities. However, it has been debated whether they are actually performing humanlike abstract reasoning or instead employing less general processes that rely on similarity to what has been seen in their training data. Here we investigate the generality of analogy-making abilities previously claimed for LLMs (Webb, Holyoak, & Lu, 2023). We take one set of analogy problems used to evaluate LLMs and create a set of "counterfactual" variants-versions that test the same abstract reasoning abilities but that are likely dissimilar from any pre-training data. We test humans and three GPT models on both the original and counterfactual problems, and show that, while the performance of humans remains high for all the problems, the GPT models' performance declines sharply on the counterfactual set. This work provides evidence that, despite previously reported successes of LLMs on analogical reasoning, these models lack the robustness and generality of human analogy-making.