Do Emotions Influence Moral Judgment in Large Language Models?

TL;DR

This study investigates how emotions influence moral judgments in large language models, revealing that positive emotions tend to increase moral acceptability while negative emotions decrease it, with effects varying by model capability.

Contribution

The paper introduces an emotion-induction pipeline for LLMs and demonstrates how emotions systematically affect moral judgments, highlighting differences from human responses.

Findings

Positive emotions increase moral acceptability in LLMs.

Negative emotions decrease moral acceptability in LLMs.

Effects can reverse binary moral judgments in up to 20% of cases.

Abstract

Large language models have been extensively studied for emotion recognition and moral reasoning as distinct capabilities, yet the extent to which emotions influence moral judgment remains underexplored. In this work, we develop an emotion-induction pipeline that infuses emotion into moral situations and evaluate shifts in moral acceptability across multiple datasets and LLMs. We observe a directional pattern: positive emotions increase moral acceptability and negative emotions decrease it, with effects strong enough to reverse binary moral judgments in up to 20% of cases, and with susceptibility scaling inversely with model capability. Our analysis further reveals that specific emotions can sometimes behave contrary to what their valence would predict (e.g., remorse paradoxically increases acceptability). A complementary human annotation study shows humans do not exhibit these systematic shifts, indicating an alignment gap in current LLMs.