# Presupposition and Reasoning in Conditionals: A Theory-Based Study of Humans and LLMs

arXiv: https://arxiv.org/abs/2605.18352 · HTML: https://mlsanity.com/paper/2605.18352/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning.
- Builds on 4 of its 46 references that are in the map.
- First posted on arXiv 18 May 2026.

## Placed in

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

## Builds on (most cited first, 4 of 4)

- [THiNK: Can Large Language Models Think-aloud?](https://mlsanity.com/paper/2505.20184/) (2025)
- [Let's CONFER: A Dataset for Evaluating Natural Language Inference Models on CONditional InFERence and Presupposition](https://mlsanity.com/paper/2506.06133/) (2025)
- [ClaimDB: A Fact Verification Benchmark over Large Structured Data](https://mlsanity.com/paper/2601.14698/) (2026)
- [Do Language Models Know Theo Has a Wife? Investigating the Proviso Problem](https://mlsanity.com/paper/2603.08358/) (2026)

## Abstract (arXiv)

Presupposition projection in conditionals is central to theories of meaning and pragmatics, yet it remains largely unevaluated in large language models. We address this gap through a parallel behavioral study comparing human judgments and LLM predictions on a normed dataset of conditional sentences that controls the relation between the antecedent and the projected presupposition. We collect likelihood ratings from 120 participants and four LLMs under matched contextual conditions. Results show that humans integrate probabilistic and pragmatic cues in their judgment, whereas LLMs show variable alignment with human patterns. Using a linguistically motivated checklist within an LLM-as-a-Judge framework, we further evaluate model reasoning. We observe models that best match human ratings often lack coherent pragmatic reasoning, while models with stronger reasoning produce less human-like judgments. These findings suggest that LLMs' performance on such tasks may result from surface pattern matching rather than pragmatic competence. Our findings highlight the importance of benchmarks grounded in linguistic theory for comparing humans and models.