Large language models show different ways to make strategic choices
The Internal Anatomy of Strategic Choice in Large Language Models
Artificial IntelligenceComputer Science and Game Theory
Summary
Large language models can act like strategic decision-makers, but the way they make choices inside their 'brain' differs. The researchers studied several models playing simple games and looked inside to see how the models used information about incentives to decide. They found that even models making similar choices might process information differently, and updating a model can change how it turns incentives into decisions without changing the choices themselves. This shows that understanding how AI makes decisions needs looking beyond just the final answers.
large language modelsstrategic choiceactivationsgame theorydecision makingincentivesmixture-of-expertsinstruction tuningmodel behaviordecodable information
Authors
Vinícius Ferraz, Leon Houf, Enrico Ferrea
Abstract
Large language models act as strategic agents and models of human choice, yet choosing like a strategic agent does not mean computing like one. We recorded activations from four open-weight models --- dense and mixture-of-experts, including a matched base--instruct pair --- in one-shot play of 144 strict ordinal $2\times2$ games. We followed a prespecified incentive from prompt, through activations, to choice. Dense models mirrored the unadjusted human decline with game complexity. Incentive and choice were detectable in every model, but models differed in whether incentive reached the choice, aligned with it and, where tested, whether strengthening it shifted preference. The base and instruction-tuned Qwen2.5 models chose almost identically at baseline yet differed in whether incentive reached choice. Fixed decision cues were distinguishable internally but changed choices selectively. Similar behaviour can rest on different computation; post-training can reshape the path from represented incentive to decision while leaving behaviour and decodable information largely intact.