Thinking effort aligns between humans and reasoning models in abductive reasoning

A new study on arXiv (2609.01867v1) finds that large reasoning models (LRMs) align with human effort in abductive reasoning, with models and humans making similar errors. The authors report that decoding methods allowing multiple reasoning paths increase alignment in reasoning cost across the three models tested.

arXiv:2609.01867v1 Announce Type: new Abstract: A major question in cognitive modeling concerns the behavioral alignment between large language models and humans across linguistic and non-linguistic tasks. Unlike standard LLMs, large reasoning models LRMs are optimized with reinforcement learning from verifiable rewards, encouraging correct solutions to reasoning tasks rather than preference-aligned responses. Recent work de Varda et al., 2025 investigates the cost of thinking in humans and LRMs by comparing human reaction times with model reasoning traces across a range of reasoning tasks. We isolate this alignment by turning to abductive reasoning: unlike deductive tasks, its difficulty cannot be inferred from formal structure and offers no shortcuts a model could exploit to mimic effort without genuine search, providing firmer ground for empirical claims of shared effort. We find further evidence of alignment between LRM and human reasoning effort, as well as evidence that models and humans tend to make similar errors. Finally, we show that decoding methods that let models explore multiple reasoning paths increase alignment in reasoning cost between humans and LRMs across the three models tested.