PhD Position on Domain Explanation for Citizens' Decision-Making

About the role

Four-year fully funded PhD at the University of Twente on domain-aware explanations that help non-expert citizens make informed decisions about complex topics (health, energy, public policy, financial choices). The work combines explainable AI, decision science, and human-computer interaction. Deadline 10 June 2026.

Responsibilities

• Build interactive interfaces that explain ML model predictions in the user’s domain language.
• Run user studies measuring comprehension, trust, and decision quality.
• Publish in CHI, IUI, ACM FAccT, and XAI venues.
• Collaborate with applied partners in Dutch public services for case-study deployments.

Requirements

• Master’s degree in computer science, cognitive science, human-computer interaction, or a closely related field.
• Background in machine learning and at least one of: HCI, explainable AI, or decision science.
• Programming proficiency in Python plus web-stack experience (JavaScript/TypeScript) for building prototypes.
• Fluent English. Dutch is welcomed for user studies but not required.

Nice to have

• Experience running quantitative user studies (controlled experiments, A/B testing).
• Prior work on explanation techniques (SHAP, LIME, counterfactuals, concept attribution).
• A first-author paper, even at workshop level.

How to apply

Open the apply link and submit CV, transcripts, motivation letter, Master’s thesis, and two reference contacts via the University of Twente application portal before 10 June 2026.