# Do LLMs "Feel"? Emotion Circuits Discovery and Control

arXiv: https://arxiv.org/abs/2510.11328 · HTML: https://mlsanity.com/paper/2510.11328/

## Key facts

- Cited by 17 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 24 references that are in the map.
- First posted on arXiv 13 Oct 2025.

## Builds on (most cited first, 11 of 11)

- [Steering Language Models With Activation Engineering](https://mlsanity.com/paper/2308.10248/) (2023)
- [MIME: MIMicking Emotions for Empathetic Response Generation](https://mlsanity.com/paper/2010.01454/) (2020)
- [ESCoT: Towards Interpretable Emotional Support Dialogue Systems](https://mlsanity.com/paper/2406.10960/) (2024)
- [Mechanistic Interpretability of Emotion Inference in Large Language Models](https://mlsanity.com/paper/2502.05489/) (2025)
- [Fine-grained Affective Processing Capabilities Emerging from Large Language Models](https://mlsanity.com/paper/2309.01664/) (2023)
- [Adapting a Language Model for Controlled Affective Text Generation](https://mlsanity.com/paper/2011.04000/) (2020)
- [Investigating Large Language Models' Perception of Emotion Using Appraisal Theory](https://mlsanity.com/paper/2310.04450/) (2023)
- [AI with Emotions: Exploring Emotional Expressions in Large Language Models](https://mlsanity.com/paper/2504.14706/) (2025)
- [Language-Specific Representation of Emotion-Concept Knowledge Causally Supports Emotion Inference](https://mlsanity.com/paper/2302.09582/) (2023)
- [AI in Mental Health: Emotional and Sentiment Analysis of Large Language Models' Responses to Depression, Anxiety, and Stress Queries](https://mlsanity.com/paper/2508.11285/) (2025)
- [The GPT-4o Shock Emotional Attachment to AI Models and Its Impact on Regulatory Acceptance: A Cross-Cultural Analysis of the Immediate Transition from GPT-4o to GPT-5](https://mlsanity.com/paper/2508.16624/) (2025)

## Built on by (most cited first, 17 of 17)

- [Emotion Concepts and their Function in a Large Language Model](https://mlsanity.com/paper/2604.07729/) (2026)
- [Good Arguments Against the People Pleasers: How Reasoning Mitigates (Yet Masks) LLM Sycophancy](https://mlsanity.com/paper/2603.16643/) (2026)
- [Discovering and Causally Validating Emotion-Sensitive Neurons in Large Audio-Language Models](https://mlsanity.com/paper/2601.03115/) (2026)
- [Inside the Unfair Judge: A Mechanistic Interpretability Account of LLM-as-Judge Bias](https://mlsanity.com/paper/2607.11871/) (2026)
- [Neuron-Level Emotion Control in Speech-Generative Large Audio-Language Models](https://mlsanity.com/paper/2603.17231/) (2026)
- [Sycophancy Suppression Can Impair Rational Updating: Anti-Sycophancy Should Preserve the Ability to Update](https://mlsanity.com/paper/2608.26511/) (2026)
- [Quantitative Introspection in Language Models: Tracking Emotive States Across Conversation](https://mlsanity.com/paper/2603.18893/) (2026)
- [Why Are Some Emotions Harder for LLMs? Uncovering the Causal Mechanisms of Emotion Inference via Sparse Autoencoders](https://mlsanity.com/paper/2604.25866/) (2026)
- [Multilingual Emotion Neurons in Large Audio-Language Models](https://mlsanity.com/paper/2608.08772/) (2026)
- [Read-Best Is Not Steer-Best: A Probing--Steering Layer Dissociation in Omni-Modal Large Language Models](https://mlsanity.com/paper/2609.22135/) (2026)
- [Understanding Affective Adaptation in Multimodal Foundation Models: Emergent Functional Specialization](https://mlsanity.com/paper/2601.15906/) (2026)
- [Mechanistic Decoding of Cognitive Constructs in Large Language Models](https://mlsanity.com/paper/2604.14593/) (2026)
- [AIPsy-Affect: A Keyword-Free Clinical Stimulus Battery for Mechanistic Interpretability of Emotion in Language Models](https://mlsanity.com/paper/2604.23719/) (2026)
- [Negative Before Positive: Asymmetric Valence Processing in Large Language Models](https://mlsanity.com/paper/2605.05653/) (2026)
- [Personalized Emotional Intelligence in Generative AI through Symbolic Affective Reasoning](https://mlsanity.com/paper/2607.10678/) (2026)
- [Emotion Across Speech and Faces: Shared Affective Mechanisms in Multimodal Foundation Models](https://mlsanity.com/paper/2608.17102/) (2026)
- [Replicating the Geometry of Emotion Representations in a Base Open-Weights Model](https://mlsanity.com/paper/2609.22208/) (2026)

## Abstract (arXiv)

As the demand for emotional intelligence in large language models (LLMs) grows, a key challenge lies in understanding the internal mechanisms that give rise to emotional expression and in controlling emotions in generated text. This study addresses three core questions: (1) Do LLMs contain context-agnostic mechanisms shaping emotional expression? (2) What form do these mechanisms take? (3) Can they be harnessed for universal emotion control? We first construct a controlled dataset, SEV (Scenario-Event with Valence), to elicit comparable internal states across emotions. Subsequently, we extract context-agnostic emotion directions that reveal consistent, cross-context encoding of emotion (Q1). We identify neurons and attention heads that locally implement emotional computation through analytical decomposition and causal analysis, and validate their causal roles via ablation and enhancement interventions. Next, we quantify each sublayer's causal influence on the model's final emotion representation and integrate the identified local components into coherent global emotion circuits that drive emotional expression (Q2). Directly modulating these circuits achieves 99.65% emotion-expression accuracy on the test set, surpassing prompting- and steering-based methods (Q3). To our knowledge, this is the first systematic study to uncover and validate emotion circuits in LLMs, offering new insights into interpretability and controllable emotional intelligence.