LLM Consumer Behavior Theory: Foundations of a Novel Research Field
Summary
This paper introduces LLM Consumer Behavior Theory, a new field studying how LLM-based autonomous agents make consumption decisions on behalf of users, formalizing preference reflection and market aggregation, and identifying open research questions.
View Cached Full Text
Cached at: 06/17/26, 05:40 AM
# LLM Consumer Behavior Theory: Foundations of a Novel Research Field Source: [https://arxiv.org/abs/2606.18005](https://arxiv.org/abs/2606.18005) [View PDF](https://arxiv.org/pdf/2606.18005) > Abstract:Large language models \(LLMs\) are increasingly deployed as autonomous agents that make consumption decisions on behalf of users\. This shift raises fundamental questions for consumer theory, which has traditionally modeled humans as the primary decision\-makers\. In this paper, we introduce LLM Consumer Behavior Theory, a new field of study concerned with analyzing consumer behavior in agentic markets\. Drawing on classical and behavioral economics alongside recent advances in Natural Language Processing, we formalize how human preferences are reflected and acted upon by LLM\-based agents, and how agent\-level decisions aggregate into market demand\. We unify previously fragmented literature on LLM decision\-making, human behavior simulation, and preference elicitation under a common economic lens, highlighting where assumptions, such as rationality and heterogeneity, may fail in agentic markets\. Rather than providing empirical validation, this paper outlines the scope of LLM consumer behavior and identifies open research questions related to alignment, preference representation, and market dynamics\. ## Submission history From: Manon Reusens \[[view email](https://arxiv.org/show-email/2ef9e29e/2606.18005)\] **\[v1\]**Tue, 16 Jun 2026 14:51:43 UTC \(5,958 KB\)
Similar Articles
Can LLMs Think Like Consumers? Benchmarking Crowd-Level Reaction Reconstruction with ConsumerSimBench
Introduces ConsumerSimBench, a benchmark for evaluating LLMs' ability to reconstruct crowd-level consumer reactions from real Chinese social media topics. Tests show frontier models cover only 47.8% of real reaction criteria, highlighting a gap between technical benchmark performance and social intuition.
Agentic Trading: When LLM Agents Meet Financial Markets
This paper presents a systematic survey and evidence map of 77 studies on LLM-based trading agents, finding that architectural experimentation is expanding rapidly but evaluation protocols, execution semantics, and reproducibility remain critical bottlenecks.
Examining Human-Like Behaviors in LLMs: A Multi-Dimensional Analysis of Model Behaviors, User Factors, and System Prompts
This paper presents a multi-dimensional analysis of human-like behaviors in LLMs, examining prevalence, effects, and controllability across 21,000 conversations from four models, finding that behaviors vary by model and user factors, with implications for responsible design.
Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems
This paper studies brand dynamics in LLM product recommendations, finding a conditional monopoly for well-known brands that can be broken by small rating advantages or authority-style marketing language, and highlighting a social dilemma in multi-brand GEO competition.
Thinking Like a Scientist? A Structural Study of LLM-Generated Research Methods
This study examines how LLMs suggest research methods (datasets, models, metrics) when prompted only with a research question, finding that LLMs exhibit a strong provider bias and propose a much narrower range of methods compared to actual papers, potentially narrowing researchers' methodological search space.