NAACL 2024
Can language model moderators improve the health of online discourse?
NAACL-HLT 2024
Affective Computing:Emotion, AI, and Human Behavior
How can AI understand and respond intelligently to human emotion?
This interdisciplinary, project-based course brings together ideas from artificial intelligence, psychology, and human-centered design to explore how emotion shapes human behavior and how computational systems can recognize, model, and influence it.
View a Previous Course Syllabus โ
Project-Based Learning
Build something that matters
Working in interdisciplinary teams, students identify a real-world problem, propose and pitch an affective computing solution, build a prototype using existing AI tools, and design an evaluation to determine whether it works.
๐ Several student projects have led to peer-reviewed publications at major conferences including NAACL, AAAI, ACII, IVA, and others.
๐ก
1. Identifya real-world problem๐ฅ
2. Forminterdisciplinary teams๐ป
3. Builda prototype with AI tools๐
4. Evaluateand measure impactBackground / Prerequisites
Students should have prior preparation through one of two routes.
๐ป Computer Science Students
Students should have substantial programming experience (e.g., CS 2201, CS 2204 or equivalent); prior coursework in artificial intelligence or machine learning is recommended.
๐ง Psychology / Cognitive Science Students
Students should have completed an introductory statistics course (e.g., PSY 2100 or PSY-PC 2110) and a course in experimental design or research methods (PSY 2150, PSY-PC 2170 or PSY-PC 2180, or equivalent). More advanced statistics coursework (e.g., PSY-PC 2120) would be beneficial.
From Course Project to Publication
Selected Publications from Student Projects
Course projects can become genuine research. These examples show how student work has developed into peer-reviewed publications.
IVA 2025
Can LLMs Generate Behaviors for Embodied Virtual Agents Based on Personality Prompting?
25th ACM International Conference on Intelligent Virtual Agents
AAAI 2026
Can LLMs Truly Embody Human Personality? Analyzing AI and Human Behavior Alignment in Dispute Resolution.
AAAI 2026
ACII 2023
Context unlocks emotions: Text-based emotion classification dataset auditing with large language models.
ACII 2023
Haptics 2024
How Visualizing Touch Can Transform Perceptions of Intensity, Realism, and Emotion?
Human Haptic Sensing and Touch Enabled Computer Applications, 2024
IVA 2011
Virtual Rapport 2.0
International Workshop on Intelligent Virtual Agents, 2011
View additional publications from course projects
- Carlos Busso, Sungbok Lee, and Shrikanth S. Narayanan. โUsing neutral speech models for emotional speech analysis.โ Interspeech, 2007.
- Jangwon Kim, Sungbok Lee, and Shrikanth S. Narayanan. โDetailed study of articulatory kinematics of critical articulators and non-critical articulators of emotional speech.โ 162nd Meeting of the Acoustical Society of America, 2011.
- Jillian Greczek, Katelyn Swift-Spong, and Maja Matariฤ. Using eye shape to improve affect recognition on a humanoid robot with limited expression. Technical report, 2011.
What Have Students Built?
Past projects have explored a wide range of applications across emotion, AI, interaction, media, and human behavior.
๐ Emotion and Speech
- End-to-end speech negotiations
- Emotionally-adapted automatic speech recognition
- Emotion Mirror
๐ฎ Games and Motivation
- Emotion-based reinforcement learning
- Regulatory fit and game performance
- Extrinsic rewards and intrinsic motivation
๐ค Social Interaction
- Real-time engagement analysis
- Emotion modeling for explanations
- Embodied virtual agents
๐ต Music and Media
- Popular music and emotion
- Empathetic music player
- Emotional Quake
๐ Perception and Sensing
- Vision-assisted animation
- Visualizing touch and haptics
- Affect recognition
โ Human-Centered AI
- AI personality and behavior generation
- Dispute resolution and human behavior
- Principled model of self-efficacy
Interested in the course?
See a previous offering for a detailed example of topics, readings, assignments, and course structure.
View a Previous Course Syllabus โ