Rigidity in LLM Bandits with Implications for Human-AI Dyads

TL;DR

The results position minimal bandits as a tractable probe of LLM decision tendencies and motivate hypotheses about how such biases could shape human-AI interaction.

Abstract

We test whether LLMs show robust decision biases. Treating models as participants in two-arm bandits, we ran 20000 trials per condition across four decoding configurations. Under symmetric rewards, models amplified positional order into stubborn one-arm policies. Under asymmetric rewards, they exploited rigidly yet underperformed an oracle and rarely re-checked. The observed patterns were consistent across manipulations of temperature and top-p, with top-k held at the provider default, indicating that the qualitative behaviours are robust to the two decoding knobs typically available to practitioners. Crucially, moving beyond descriptive metrics to computational modelling, a hierarchical Rescorla-Wagner-softmax fit revealed the underlying strategies: low learning rates and very high inverse temperatures, which together explain both noise-to-bias amplification and rigid exploitation. These results position minimal bandits as a tractable probe of LLM decision tendencies and motivate hypotheses about how such biases could shape human-AI interaction.

Figures (5)

• Figure 1: Symmetric Bandit Behavioural Metrics
• Figure 2: Asymmetric Bandit Behavioural Metrics
• Figure 3: Posterior Densities of Group‑level Parameters
• Figure 4: Test-retest Reliability on the Symmetric Bandit
• Figure 5: Test-retest Reliability on the Asymmetric Bandit