Prof. Dr. Catholijn M. Jonker


    Interactive Intelligence
    Explainable Artificial Intelligence
    Negotiation & Deliberation & Decision making

How to reach me

Prof. Dr. C.M. Jonker
Interactive Intelligence Group

Dept. of Intelligent Systems
Faculty of Electrical Engineering, Mathematics & Computer Science

Delft University of Technology

  • Room: W6.600
  • Telephone: +31.15.2781315
  • Mobile phone: +31.6.48875207
Prof. Dr. C.M. Jonker
Explainable Artificial Intelligence

Dept. of Media Technology
Faculty of Science
The Leiden Institute of Advanced Computer Science (LIACS)

Leiden University

    • Room: 128
    • Telephone: +31.71.5277061
    • Mobile phone: +31.6.48875207


My research interests are the modelling and simulation of cognitive processes and concepts such as trust, negotiation, teamwork and the dynamics of individual agents and organisations. I enjoy working in interdisciplinary teams and creating synergy between humans and technology by understanding, shaping and using fundamentals of intelligence and interaction. I am inspired by social intelligence theories, and, e.g., use concepts such as social practice to improve the interactive intelligence of agents.

Delft University of Technology is my work base since 2006. It is a great university that breaths technology and provides me with fantastic colleagues. Read why I love working there, especially with the Interactive Intelligence group.

Universiteit Leiden has employed me as a part-time professor since 2017. It is a great university that loves technology but loves other scientific disciplines just as much. It offers me an opportunity to connect to those other disciplines in a way that is not possible at TU Delft.

My current main research projects are:

Negotiation Support Systems for which I received a (NWO-STW VICI grant) in 2008. With this ongoing project I aim to support human negotiators. Over time various fantastic people have joined the team for some years. With them I study ways to elicit information from negotiation parties about their preference profiles and underlying concerns. Qualitative reasoning about the profiles and concerns is studied as that should make it easier to explain the reasoning to the human user. As emotions sometimes play an important role in negotiation, a training environment has been created to learn how to deal with emotions in negotiations. The negotiation environment GENIUS is used top groups in the world for developing the best negotiation strategies in automated negotiations.

VRAPT – Virtual Reality Aggression Prevention Training is a project to that focuses on building a Virtual Reality therapy system that facilitates treatment. The patients have to interact with avatars in a virtual world. Our role in the project is part of the effort to make the experience real. To increase presence, the avatar has to respond appropriately and fast. This is facilitated by two main components: behaviour recognition and behaviour generation. We focus on behaviour recognition. Multiple modalities are used for the behaviour recognition component, such as audio, kinect and physiological data, to predict negative arousal and aggressive tendencies.

Socially Adaptive Electronic Partners (SAEP). The overall aim of this VIDI project of Birna van Riemsdijk is to develop a reasoning framework that combines logic and quantitative techniques for Socially Adaptive Electronic Partners (SAEPs) that adapt their behavior to norms and values of people. This becomes more and more important as technology becomes an integral part of our daily lives. The computational reasoning techniques are aimed at determining when and to what extent norm-compliance can be guaranteed, and deciding what to do if in exceptional situations a norm cannot or should not be complied with. We are working on this project together with Pietro Pasotti by studying how to help people with their habits and habit formation.

Sensing Heterogeneous Information Network Environment (SHINE). Research theme: How can heterogeneous resources (people, mobile sensors, fixed sensors, social media, information systems, …) self-organize for answering information needs? My contribution is in the field of negotiation and of organisation-aware agents.

Smart M2M Grids the M2M Internet for dynamic M2M Information Business ecosystem. Our contribution in that project focuses on multi-party negotation, where we study protocols, bidding and search heuristics and human-in-the-loop automated negotations.

IQmulus which is a EU project to create a High-volume Fusion and Analysis Platform for Geospatial Point Clouds, Coverages and Volumetric Data Sets. Our contribution focuses on human-computer interaction. We employ value-sensitive design methods and focus group interviews to support the design of IQmulus.

Shared Mental Models is a project that studies what human and artificial actors should know about each other to get the best experience and task performance. A research questions is, how to formalize that knowledge in what is known as shared mental models. A related question is how to quantifiably measure sharedness and develop agents that work to improve sharedness.

Research hobby

Dog cognition is a topic of research that arose from my passion for dogs and science. When my trainer at De Roedel said he could predict how my dog would perform a task based on paw the dog departed with to perform the task, he sparked the unquenchable thirst to see scientific evidence and to find out more.

We accept self-funded research trainees to help us with this research. If you are interested in an alien intelligence that has been living close to us humans for many thousands of years and you have an aptitude for formalisations and simulations, please contact>me.

Application areas

Over the years I learned most about the following application areas

  • Negotiation
  • Trade
  • Animal behaviour
  • Lifestyle and Personalized Care
  • Cybernetic Incident Management & International Peace Keeping Missions


Civic tech project PoliFLW helpt bij het volgen van duizenden lokale politici.

Terwijl op lokale volksvertegenwoordigers een zware verantwoordelijkheid rust doordat gemeenten er steeds meer taken bij krijgen, kampt de journalistiek met te weinig middelen en tijd om deze lokale politici goed te kunnen volgen. Digitale hulpmiddelen kunnen daarbij helpen. Open State Foundation lanceert daarom vandaag het civic tech project PoliFLW.

Met PoliFLW kost het journalisten, lokale belangenbehartigers en minder tijd om de uitlatingen van politieke partijen en politici te volgen. Hiermee helpen we journalisten om grote groepen politici, waaronder meer dan tienduizend lokale volksvertegenwoordigers die normaal niet in beeld waren, gelijktijdig te volgen in wat zij in het openbaar zeggen.

Lees verder op:

 PoliFLW is een systeem ontwikkeld in samenwerking met afstudeerders informatica Joost Rothweiler en Max van Zoest van de Technische Universiteit Delft, onder begeleiding van professor Catholijn Jonker.



Luister naar de meningen van:

Wetenschapsspecialist Bennie Mols, gespecialiseerd in kunstmatige intelligentie

Aske Plaat, hoogleraar Data Science aan Universiteit Leiden

Catholijn Jonker, hoogleraar Explainable A.I. in Leiden en Interactive Intelligence aan de TU Delft


Opening academic year Delft University of Technology 2017-2018


In Delta, onafhankelijk universiteitsblad TUDelft


Women in Technology

We need more women in technology, we need more women in the higher academic ranks! Why? Because of their talent! How many? Proportional to the number of women in our society! Is it only about women? No, it is about diversity in LGBT, race, and all kinds of other aspects that characterize people in ways they can’t consciously change. I just happen to be a woman, and for women academic life is unnecessarily tougher than for men. So I decided spend most my efforts for diversity on women in science.

In de volkskrant



For all publication please follow this link

Boards & Committees

(Board)memberships of organisations:

the International Joint Conference in Autonomous Agents and Multi-Agent Systems (AAMAS) conferences:

International Joint Conference on Artificial Intelligence (IJCAI)

Programme chair of IEA/AIE 2013

Teaching & PhD Students

I am usually involved in the following courses (more information on Blackboard):

  • IN4010TU Artificial Intelligence Techniques
  • TI1906 Logic Based Artificial Intelligence

In the past I have (co)-taught many courses on topics such as multi-agent systems, non-monotonic logics, temporal logics, artificial intelligence, and negotiation.

Current Ph.D. students

  1. Anouk Nijs, started Sept 2018, NWO Perspectief Programma Citius, Altius, Sanius. Promotors: Prof. dr. Peter Beek (VU), and Prof. dr. CM Jonker.
  2. Ralf Wolter, started February 2018, Explainable AI for Reception Robots. Supervisory team: Dr. Koen Hindriks  and Prof. dr. CM Jonker (promotors).
  3. Nanja Smets (guest) works on the MECA project, supervision together with Prof. Dr. Mark Neerincx. Her project is related to the Shared Mental Models project.
  4. Vincent Koeman. working on self-explaining software agents. Started November 2014. Co-supervised by Dr. Koen Hindriks.
  5. Ilir Kola. Computational reasoning tools as part of the CoreSAEP project. Started December 2017. Co-supervised by Birna van Riemsdijk.
  6. Thomas Moerland. Affective Computing and Reinforcement Learning. Started November 2015. Co-supervised by Joost Broekens.
  7. Rijk Mercuur. Rijk is working on social practices and nervousnet. Intelligent agents should understand social practices to work together (and with humans) through Nervousnet in an effective and socially acceptable way. Rijk’s daily supervisor is Virginia Dignum.
  8. Xueling Li is working on Objects with Intent. The supervisory team consists of Marco Rozendaal (daily supervisor), Kaspar Jansen, and myself.


List of students who succesfully defended their thesis

Attitude is everything, so pick a good one


I am a computer scientist researching and teaching in artificial intelligence, and a bit about dog cognition. My family, friends and dog have my devotion. Reading is a long-life addiction for me, with a preference for scifi and fantasy. Sculpting in clay and doing heavy garden work is my way of reducing brain activity. American Kenpo Karate has taught me some much needed lessons of how to deal with human behaviour. Dogs and horses have taught me some much needed lessons about myself and inspire me in my work towards artificial intelligence and in particular to interactive intelligence.


I have much to thank for to dogs, and in particular the Swiss Mountain Dogs (Appenzeller Sennenhunde), and all the other dogs that became my friends and which I had the privilege to walk through life with for some time. Visit this page to meet some of my friends at least in spirit.

  • Eppy and me. Hey, we are just getting to know each other. Love you already.


  • Artos and me. Thank you, Artos, for all you gave me and our friends, so sad when you passed away end of 2014. You will always be the gentleman to me and my deepest respect for your stability in all situations. Our best times were spent in Zeeuws-Vlaanderen and with De Roedel. I will never forget that you hid objects for us to find.


  • Igor: my first dog. Thank you, Igor, for all the fun we had. You are still in my thoughts, even though it is already so long ago that you passed away (1999). Kasteel Groeneveld and the beaches of Zeeuws-Vlaanderen will always be the places that I remember you having the most fun.

Clay sculpting

The touch and feel of clay has the power to take me out of my usual though patterns and gives me the opportunity to connect reason with observations. Get a glimpse of my clay projects.

Clay sculpting


Another way of leaving reality is to retreat into fantasy by reading and imagining other worlds and what it might be like to be someone else. Let me introduce you to some of my favorite authors and books.

American Kenpo Karate

Martial Arts have played a role in my life from early childhood. I was introduced to the martial arts through Judo which my parents insisted I do before allowing me to ride horses. The point being not break any bones when falling to the ground (still very valid now). Find out how much more it has come to mean to me over the years.


Horses are the second non-human species that have taught me so much about myself, but also about other humans and even something about dogs. Oh, and about animal cognition in general. Let me share it with you.


Contemporary societies need new approaches to enable mass participation in support of future resilient democracies. High-quality deliberation and broad participation have been deemed mutually incompatible, as all attempts to combine these democratic processes have failed. Social Media platforms facilitate mass engagement in societal problems, but deliberation on such platforms is dysfunctional due to issues such as trolling. Existing approaches using mini-publics for deliberation (small representative groups of randomly selected citizens) by definition segregate deliberation from the masses. The PLATO project aims to create synergy between humans and artificial agents to bridge the gap between democratic methods of deliberation and mass participation. The key concept is collaborative agent technology. These intelligent agents deploy machine learning, argumentation support systems, deliberative methods, social media platforms and deliberation space exploration techniques to monitor democratic processes. They will inform and consult the public about new insights and problems, thus enriching deliberation and engaging a broader cross-section of the public. To further enhance deliberation quality, we study deliberation space exploration methods to inject deliberations with new and better potential solutions. read more

Existing negotiation support systems are based almost exclusively on quantitative models of preferences. It is however difficult to elicit such models from users, while complex utility functions also have serious computational limitations. Qualitative preference models in many regards would provide a better correspondence with human cognitive representations. This would allow a human user to more naturally interact with a machine negotiating on his/her behalf. The candidate’s research will focus on the design of an expressive qualitative preference language and associated techniques for using this language in negotiation. The techniques that need to be developed concern the elicitation of qualitative preferences, the reasoning with possibly incomplete qualitative preferences, and associated strategies for negotiation. The research has been performed in close collaboration with other sub-projects with the aim of developing a negotiation support system. read more