The Riddle Riddle: Testing Flexible Reasoning in Large Language Models and Humans

• URL: http://arxiv.org/abs/2606.27103v1
• Date: Thu, 25 Jun 2026 14:41:12 GMT
• Title: The Riddle Riddle: Testing Flexible Reasoning in Large Language Models and Humans
• Abstract summary: Humans flexibly adapt their reasoning strategies to the requirements of a given problem.<n>Riddle riddles are word problems written to mimic popular riddles, but altered so their answers only require literal interpretations.<n>We show humans and large language models fail on this paradigm in opposite directions.
• Score: 5.719242751508825
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Humans flexibly adapt their reasoning strategies to the requirements of a given problem. Large language models (LLMs) have performed well on many cognitive tasks, however, it is unclear whether this accuracy is a result of pattern matching from training data or flexible reasoning. Here, we introduce a novel paradigm to test this question: the riddle riddle paradigm. Riddle riddles are word problems written to mimic popular riddles, but altered so their answers only require literal interpretations. Identifying correct answers requires looking past the structure of each question and flexibly apply different reasoning strategies based on the content. If LLMs respond to surface features, such as form, a riddle-like structure should cause models to use an inventive reasoning strategy even when a literal interpretation suffices. Alternatively, if LLMs reason based on content, they should flexibly switch strategies when appropriate. Across two experiments with nine state-of-the-art LLMs and 100 human participants, we show humans and LLMs fail on this paradigm in opposite directions. LLMs were far more accurate on genuine riddles than on riddle riddles (84.9% vs. 50.7%); whereas humans showed the reverse effect (50.5% vs. 80.5%). Error analysis shows that 90.8% of LLM errors on riddle riddles (the condition where they show diminished performance) were due to inappropriate use of inventive reasoning while only 57.6% of human errors on genuine riddles were due to overextending literal reasoning. Thus, while both groups make mistakes, reasoning mistakes are made more often by LLMs than by humans. Overall, LLMs' strong performance on genuine riddles may reflect memory retrieval rather than flexible strategy selection, and without stimuli designed to elicit this contrast, it becomes easy to conflate LLM-generated outputs that look like reasoning with genuine reasoning.

Related papers

• Intermediate Languages Matter: Formal Choice Drives Neurosymbolic LLM Reasoning [50.99811144731619]
  We show that the choice of formal language affects both the syntactic and the semantic reasoning capability.<n>We conclude that on average, context-aware encodings help LLMs to reason, while there is no apparent effect of using comments or markdown syntax.
  arXiv  Detail & Related papers  (2025-02-24T14:49:52Z)
• Empowering LLMs with Logical Reasoning: A Comprehensive Survey [49.91445266392609]
  Large language models (LLMs) have achieved remarkable successes on various tasks.<n>Recent studies have found that there are still significant challenges to the logical reasoning abilities of LLMs.
  arXiv  Detail & Related papers  (2025-02-21T18:20:35Z)
• No Need for Explanations: LLMs can implicitly learn from mistakes in-context [15.70441204657966]
  We study why Large Language Models learn from mistakes more effectively without explicit corrective rationales.<n>We find evidence that, while incorrect answers are more beneficial for LLM learning, explicit corrective rationales over-constrain the model.
  arXiv  Detail & Related papers  (2025-02-12T16:31:21Z)
• On Memorization of Large Language Models in Logical Reasoning [70.94164038947078]
  Large language models (LLMs) achieve good performance on challenging reasoning benchmarks, yet could also make basic reasoning mistakes.<n>One hypothesis is that the increasingly high and nearly saturated performance could be due to the memorization of similar problems.<n>We show that fine-tuning leads to heavy memorization, but it also consistently improves generalization performance.
  arXiv  Detail & Related papers  (2024-10-30T15:31:54Z)
• Do Large Language Models Truly Grasp Mathematics? An Empirical Exploration From Cognitive Psychology [15.572185318032139]
  We show that, even with the use of Chains of Thought prompts, mainstream LLMs have a high error rate when solving modified CRT problems.<n>Specifically, the average accuracy rate dropped by up to 50% compared to the original questions.<n>This finding challenges the belief that LLMs have genuine mathematical reasoning abilities comparable to humans.
  arXiv  Detail & Related papers  (2024-10-19T05:01:56Z)
• Not All LLM Reasoners Are Created Equal [58.236453890457476]
  We study the depth of grade-school math problem-solving capabilities of LLMs. We evaluate their performance on pairs of existing math word problems together.
  arXiv  Detail & Related papers  (2024-10-02T17:01:10Z)
• Order Matters in Hallucination: Reasoning Order as Benchmark and Reflexive Prompting for Large-Language-Models [0.0]
  Large language models (LLMs) have generated significant attention since their inception, finding applications across various academic and industrial domains.<n>LLMs often suffer from the "hallucination problem", where outputs, though grammatically and logically coherent, lack factual accuracy or are entirely fabricated.
  arXiv  Detail & Related papers  (2024-08-09T14:34:32Z)
• Logic Query of Thoughts: Guiding Large Language Models to Answer Complex Logic Queries with Knowledge Graphs [102.37496443389203]
  'Logic-Query-of-Thoughts' (LGOT) is the first of its kind to combine knowledge graph reasoning and large language models.<n>Our experimental findings demonstrate substantial performance enhancements, with up to 20% improvement over ChatGPT.
  arXiv  Detail & Related papers  (2024-03-17T17:01:45Z)
• Conditional and Modal Reasoning in Large Language Models [1.999925939110439]
  We focus on inference patterns involving conditionals and modals. All the LLMs we tested make some basic mistakes with conditionals or modals. Even the best performing LLMs make basic errors in modal reasoning.
  arXiv  Detail & Related papers  (2024-01-30T16:56:54Z)
• LogicAsker: Evaluating and Improving the Logical Reasoning Ability of Large Language Models [63.14196038655506]
  We introduce LogicAsker, a novel approach for evaluating and enhancing the logical reasoning capabilities of large language models (LLMs) Our methodology reveals significant gaps in LLMs' learning of logical rules, with identified reasoning failures ranging from 29% to 90% across different models. We leverage these findings to construct targeted demonstration examples and fine-tune data, notably enhancing logical reasoning in models like GPT-4o by up to 5%.
  arXiv  Detail & Related papers  (2024-01-01T13:53:53Z)
• Can ChatGPT Defend its Belief in Truth? Evaluating LLM Reasoning via Debate [19.887103433032774]
  Large language models (LLMs) have shown impressive performance in complex reasoning tasks. This work explores testing LLMs' reasoning by engaging with them in a debate-like conversation. We find that despite their impressive performance, LLMs like ChatGPT cannot maintain their beliefs in truth for a significant portion of examples.
  arXiv  Detail & Related papers  (2023-05-22T15:47:31Z)
• Logic-LM: Empowering Large Language Models with Symbolic Solvers for Faithful Logical Reasoning [101.26814728062065]
  Large Language Models (LLMs) have shown human-like reasoning abilities but still struggle with complex logical problems. This paper introduces a novel framework, Logic-LM, which integrates LLMs with symbolic solvers to improve logical problem-solving.
  arXiv  Detail & Related papers  (2023-05-20T22:25:38Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.