Stable Behavior, Limited Variation: Persona Validity in LLM Agents for Urban Sentiment Perception

TL;DR

This study examines whether persona prompts in multimodal LLMs lead to meaningful, reproducible variations in urban sentiment perception, finding limited impact of personas and questioning their utility.

Contribution

It provides empirical evidence that simple persona prompting in LLMs results in stable but limited behavioral variation in urban sentiment analysis.

Findings

Agents show strong within-persona consistency.

Cross-persona variation is statistically detectable but practically modest.

Performance drops with finer sentiment granularity.

Abstract

Large Language Models (LLMs) are increasingly used as proxies for human perception in urban analysis, yet it remains unclear whether persona prompting produces meaningful and reproducible behavioral diversity. We investigate whether distinct personas influence urban sentiment judgments generated by multimodal LLMs. Using a factorial set of personas spanning gender, economic status, political orientation, and personality, we instantiate multiple agents per persona to evaluate urban scene images from the PerceptSent dataset and assess both within-persona consistency and cross-persona variation. Results show strong convergence among agents sharing a persona, indicating stable and reproducible behavior. However, cross-persona differentiation is limited: economic status and personality induce statistically detectable but practically modest variation, while gender shows no measurable effect and political orientation only negligible impact. Agents also exhibit an extremity bias, collapsing intermediate sentiment categories common in human annotations. As a result, performance remains strong on coarse-grained polarity tasks but degrades as sentiment resolution increases, suggesting that simple label-based persona prompting does not capture fine-grained perceptual judgments. To isolate the contribution of persona conditioning, we additionally evaluate the same model without personas. Surprisingly, the no-persona model sometimes matches or exceeds persona-conditioned agreement with human labels across all task variants, suggesting that simple label-based persona prompting may add limited annotation value in this setting.