Comparative reversal learning reveals rigid adaptation in LLMs under non-stationary uncertainty

Research reveals that large language models like DeepSeek-V3.2, Gemini-3, and GPT-5.2 show rigid adaptation patterns when learning from changing environments, particularly struggling with loss-based learning compared to humans. The study found LLMs demonstrate asymmetric responses to positive versus negative feedback, with some models showing extreme perseveration after environmental changes.

• →LLMs show asymmetric learning patterns with strong win-stay behavior but weak lose-shift responses compared to humans.
• →DeepSeek-V3.2 demonstrated the most rigid adaptation with extreme perseveration after environmental reversals.
• →Gemini-3 and GPT-5.2 adapted more rapidly than DeepSeek but still remained less loss-sensitive than human benchmarks.
• →High aggregate performance can coexist with rigid adaptation patterns in volatile environments.
• →Research identifies specific mechanisms behind AI rigidity including weak loss learning and inflated policy determinism.