Summary
The study looked at whether big language models, which are computer programs that understand and generate text, can judge people's morals like humans do. The authors found that these models can rank a whistleblower's character similarly to humans but see their reasons differently, thinking whistleblowers are more helpful and less self-interested. The models also didn't change their views much when given more context about the whistleblower. This means that even if models agree on overall moral ratings, they may not fully capture the complex reasons behind people's judgments. The authors suggest checking how models respond in detail rather than just comparing average scores.
large language modelswhistleblowermoral judgmentmotive attributionpsychological researchsimulationhelpfulnessself-interestcontext sensitivityresponse patterns
Authors
Xiaoyan Wu, Jean-Claude Dreher
Abstract
Large language models (LLMs) are used to simulate human participants in psychological research. We asked whether LLMs that reproduce human evaluations of a whistleblower's moral character also reproduce the motive attributions that accompany them. Five LLMs and two human samples (N = 125 and N = 742) evaluated a physician who either remained silent about fraudulent billing or reported it to a hospital, regulator, or newspaper. Models reproduced the human ranking of the physician's moral character but portrayed whistleblowers as more helpful, less self-interested, and less hostile. In four of five models, competitive motives were less strongly associated with moral-character judgments. Model ratings changed little when prompts reproduced the narratives and demographic profiles of both human samples, although this comparison cannot isolate a perspective effect. Thus, agreement in average ratings can conceal differences in attributed motives, relationships among judgments, and sensitivity to context. Validating LLMs as simulated participants therefore requires testing psychologically informative response patterns, not average agreement alone.