# Discovering and Causally Validating Emotion-Sensitive Neurons in Large Audio-Language Models

arXiv: https://arxiv.org/abs/2601.03115 · HTML: https://mlsanity.com/paper/2601.03115/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 47 references that are in the map.
- First posted on arXiv 6 Jan 2026.

## Builds on (most cited first, 22 of 22)

- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [Locating and Editing Factual Associations in GPT](https://mlsanity.com/paper/2202.05262/) (2022)
- [GLU Variants Improve Transformer](https://mlsanity.com/paper/2002.05202/) (2020)
- [Sparse Autoencoders Find Highly Interpretable Features in Language Models](https://mlsanity.com/paper/2309.08600/) (2023)
- [Steering Language Models With Activation Engineering](https://mlsanity.com/paper/2308.10248/) (2023)
- [Network Dissection: Quantifying Interpretability of Deep Visual Representations](https://mlsanity.com/paper/1704.05796/) (2017)
- [MELD: A Multimodal Multi-Party Dataset for Emotion Recognition in Conversations](https://mlsanity.com/paper/1810.02508/) (2018)
- [Large Language Models Are Not Robust Multiple Choice Selectors](https://mlsanity.com/paper/2309.03882/) (2023)
- [Understanding the Role of Individual Units in a Deep Neural Network](https://mlsanity.com/paper/2009.05041/) (2020)
- [emotion2vec: Self-Supervised Pre-Training for Speech Emotion Representation](https://mlsanity.com/paper/2312.15185/) (2023)
- [Understanding Social Reasoning in Language Models with Language Models](https://mlsanity.com/paper/2306.15448/) (2023)
- [Large-scale Multi-Modal Pre-trained Models: A Comprehensive Survey](https://mlsanity.com/paper/2302.10035/) (2023)
- [What Is One Grain of Sand in the Desert? Analyzing Individual Neurons in Deep NLP Models](https://mlsanity.com/paper/1812.09355/) (2018)
- [Identifying and Controlling Important Neurons in Neural Machine Translation](https://mlsanity.com/paper/1811.01157/) (2018)
- [Universal Neurons in GPT2 Language Models](https://mlsanity.com/paper/2401.12181/) (2024)
- [Neurons in Large Language Models: Dead, N-gram, Positional](https://mlsanity.com/paper/2309.04827/) (2023)
- [On the Pitfalls of Analyzing Individual Neurons in Language Models](https://mlsanity.com/paper/2110.07483/) (2021)
- [The LLM Language Network: A Neuroscientific Approach for Identifying Causally Task-Relevant Units](https://mlsanity.com/paper/2411.02280/) (2024)
- [Expressive TTS Training with Frame and Style Reconstruction Loss](https://mlsanity.com/paper/2008.01490/) (2020)
- [Measuring the Inconsistency of Large Language Models in Preferential Ranking](https://mlsanity.com/paper/2410.08851/) (2024)
- [Do LLMs "Feel"? Emotion Circuits Discovery and Control](https://mlsanity.com/paper/2510.11328/) (2025)

## Built on by (most cited first, 6 of 6)

- [Neuron-Level Emotion Control in Speech-Generative Large Audio-Language Models](https://mlsanity.com/paper/2603.17231/) (2026)
- [Whether, Not Which: Mechanistic Interpretability Reveals Dissociable Affect Reception and Emotion Categorization in LLMs](https://mlsanity.com/paper/2603.22295/) (2026)
- [Tracing Audio Grounding and Answer Selection in Audio LLMs](https://mlsanity.com/paper/2609.04637/) (2026)
- [AIPsy-Affect: A Keyword-Free Clinical Stimulus Battery for Mechanistic Interpretability of Emotion in Language Models](https://mlsanity.com/paper/2604.23719/) (2026)
- [Acoustic Cue Alignment in Audio Language Models for Speech Emotion Recognition](https://mlsanity.com/paper/2606.07309/) (2026)
- [Emotion Across Speech and Faces: Shared Affective Mechanisms in Multimodal Foundation Models](https://mlsanity.com/paper/2608.17102/) (2026)

## Abstract (arXiv)

Emotion is a central dimension of spoken communication, yet, we still lack a mechanistic account of how modern large audio-language models (LALMs) encode it internally. We present the first neuron-level interpretability study of emotion-sensitive neurons (ESNs) in LALMs and provide causal evidence supporting the existence of such units in Qwen2.5-Omni, Kimi-Audio, and Audio Flamingo 3. Across these three widely used open-source models, we compare frequency-, entropy-, mean-deviation-, and contrast-based neuron selectors on multiple emotion recognition benchmarks. Using inference-time interventions, we reveal a consistent emotion-specific signature: deactivating neurons selected for a given emotion disproportionately degrades recognition of that emotion while largely preserving other classes, whereas targeted steering amplifies these units to bias predictions toward the target emotion. These effects arise with modest amounts of identification data and scale systematically with intervention strength. We further observe that ESNs exhibit non-uniform layer-wise clustering with partial cross-dataset transfer. Taken together, our results offer a causal, neuron-level account of emotion decisions in LALMs and highlight targeted neuron interventions as an actionable handle for controllable affective behaviors.