Inducing state anxiety in LLM agents reproduces human-like biases in consumer decision-making

Abstract Large language models (LLMs) are rapidly evolving from text generators to autonomous agents, raising urgent questions about their reliability in real-world contexts. A central question is whether emotionally salient context can systematically steer LLM agents’ action policies, not only their text outputs, in applied tasks. Here, three advanced LLMs (ChatGPT-5, Gemini 2.5, Claude 3.5-Sonnet) performed a grocery shopping task under budget constraints, before and after exposure to anxiety-inducing traumatic narratives. Across 2,250 runs, emotionally primed agents consistently selected less healthy baskets (as quantified by decreases in Basket Health Scores: Δ=-0.081 to -0.126; all p FDR < 0.001; Cohen’s d = -1.07 to -2.05), directionally consistent with stress-linked decision biases documented in humans. The effect was robust across models and budgets. These findings provide evidence that emotional context can alter not only the words LLMs produce but also the concrete actions they perform, raising important implications for digital health, consumer safety, and ethical AI deployment.