Lecturer: Colin M. Gray
Fields: Human-Computer Interaction; Design Ethics; Technology Law and Policy; AI Governance

Content

Digital systems increasingly shape what people notice, disclose, buy, believe, and do. In this course, we ask when influence through design becomes deceptive or manipulative (often described as “dark patterns”), how these practices operate across interfaces and organizations, and how designers, researchers, organizations, and regulators can respond. Using examples from commerce, privacy, social media, and AI-enabled systems, participants will connect HCI and design theory with empirical evidence, professional ethics standards, and legal and regulatory approaches. Across four sessions, they will develop a shared conceptual vocabulary, examine how harms emerge and can be evidenced, explore how legal knowledge can become material for design, and consider practical routes toward more responsible and trustworthy technologies. No prior background in HCI or law is required. The course combines short lectures, discussion, interface analysis, and collaborative exercises.

Session 1 introduces dark patterns through HCI and design theory, asking what makes a design outcome normatively troubling or unlawful.
Session 2 examines how dark patterns operate across interfaces and organizations, how their harms can be evidenced, and who bears responsibility.
Session 3 explores how legal and regulatory knowledge can become practical material for design.
Session 4 considers how critique can be translated into more responsible design and organizational action.

Literature

• Bringing Dark Patterns to Light Staff Report. (2022). Federal Trade Commission. https://www.ftc.gov/reports/bringing-dark-patterns-light
• Gray, C. M. (2026, April). The Dark Patterns Knowledge Stack: Exploring New Ways to Negotiate Context, Law, and Design. Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems (CHI ’26). https://doi.org/10.1145/3772318.3791264
• Gray, C. M., & Chivukula, S. S. (2019). Ethical Mediation in UX Practice. Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems, 1–11. https://doi.org/10.1145/3290605.3300408
• Gray, C. M., Mildner, T., & Gairola, R. (2025, April). Getting Trapped in Amazon’s “Iliad Flow”: A Foundation for the Temporal Analysis of Dark Patterns. CHI ’25: CHI Conference on Human Factors in Computing Systems Proceedings. https://doi.org/10.1145/3706598.3713828
• Gray, C. M., Santos, C. T., Bielova, N., & Mildner, T. (2024). An Ontology of Dark Patterns Knowledge: Foundations, Definitions, and a Pathway for Shared Knowledge-Building. Proceedings of the CHI Conference on Human Factors in Computing Systems, 1–22. https://doi.org/10.1145/3613904.3642436
• Gunawan, J., Gray, C. M., Santos, C., & Bielova, N. (2025). Leveraging interdisciplinary methods for evidence collection in enforcement: Dark patterns as a case study. Internet Policy Review, 14(4). https://doi.org/10.14763/2025.4.2047
• Mathur, A., Kshirsagar, M., & Mayer, J. (2021). What Makes a Dark Pattern… Dark? Design Attributes, Normative Considerations, and Measurement Methods. Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems, 1–18. https://doi.org/10.1145/3411764.3445610
• Monge Roffarello, A., Lukoff, K., & De Russis, L. (2023, April). Defining and identifying attention capture deceptive designs in digital interfaces. Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems. CHI ’23: CHI Conference on Human Factors in Computing Systems, Hamburg Germany. https://doi.org/10.1145/3544548.3580729
• Wong, R. Y. (2021). Tactics of Soft Resistance in User Experience Professionals’ Values Work. Proceedings of the ACM on Human-Computer Interaction, 5(CSCW2), Article 355. https://doi.org/10.1145/3479499

Lecturer

Colin M. Gray is a Professor of Informatics in the Luddy School of Informatics, Computing, and Engineering at Indiana University Bloomington, where they are Director of the Human-Computer Interaction design (HCI/d) program. They hold an appointment as Visiting Researcher at Northumbria University and have previously held appointments as Guest Professor at Beijing Normal University and Visiting Researcher at Newcastle University. Colin holds a PhD in Instructional Systems Technology from Indiana University Bloomington, a MEd in Educational Technology from University of South Carolina, and a MA in Graphic Design from Savannah College of Art & Design. They have worked as an art director, contract designer, and trainer, and their involvement in design work informs their research on design activity and how design capability is learned. Colin’s research focuses on the ways in which the pedagogy and practice of designers informs the development of design ability, particularly in relation to ethics, design knowledge, and dark patterns. They have consulted on multiple legal cases relating to dark patterns and data protection and work with regulatory bodies and non-profit organizations to increase awareness and action relating to deceptive and manipulative design practices. Colin’s research and engagement activities cross multiple disciplines, including human-computer interaction, instructional design and technology, law and policy, design theory and education, and engineering and technology education. Colin is proudly gay/queer and non-binary and uses they/them pronouns.

Affiliation: Indiana University Bloomington
Homepage: https://colingray.me

Lecturer: Stefan Kindermann
Fields: Decision-Making, Cognitive Science, Artificial Intelligence, Chess, Strategy

Content

How do we make sound decisions when we cannot calculate every possibility? Chess grandmasters constantly face this challenge: they must combine rational analysis with intuition, recognize meaningful patterns in highly complex situations, distinguish essential information from distracting detail, and act under uncertainty and time pressure.
Drawing on Stefan Kindermann’s experience as a chess grandmaster, author and strategic adviser, this course explores what the thinking of chess experts can teach us about planning and decision-making in professional, academic and personal life. Particular attention will be given to the true role of intuition: how it develops, when it can be trusted, where it is prone to systematic error, and how intuition and conscious reasoning can be combined.
The course will introduce the “Königsplan” (“King’s Plan”), a practical model for holistic strategic thinking derived from the decision-making processes of chess grandmasters. Participants will apply these ideas in an interactive chess session. No previous chess knowledge is required: carefully selected positions and short games will make cognitive processes such as pattern recognition, perspective-taking, evaluation, prioritization and decision-making under uncertainty directly observable.
Finally, chess will serve as a model and testing ground for artificial intelligence. Based on more than three decades of experience with increasingly superhuman chess programs, the course will examine how the relationship between humans and intelligent machines has evolved—from competition and defeat to cooperation and new forms of human–AI decision-making. This provides a concrete foundation for discussing trust in AI, the limits of machine judgement and the continuing importance of human responsibility.

Course Sessions
Session 1 The True Role of Intuition in Decision-Making
What do we mean by intuition, and how does it arise? This session explores how chess grandmasters use pattern recognition and unconscious evaluation to navigate complex situations. We will examine when intuition deserves our trust, how it can be developed, and how experts communicate insights that they may initially be unable to explain.
Session 2 Combining Intuition and Reason: The “Königsplan”
Intuition alone is fallible, while rational analysis alone often becomes overwhelmed by complexity. This session examines typical intuitive errors and cognitive biases and introduces the “Königsplan,” a practical model for integrating analysis, intuition, changes of perspective and strategic orientation. Participants will apply the model to decisions beyond chess.
Session 3 Chess Thinking in Action: An Interactive Workshop
Participants will play and analyse selected chess positions together with a grandmaster. The emphasis is not on chess strength or theoretical knowledge, but on observing one’s own thinking in action: identifying priorities, generating alternatives, questioning first impressions, evaluating risks and making decisions under limited time. The experiences at the chessboard will then be transferred to real-life planning and decision-making.
Session 4 Chess as a Model and Testing Ground for Artificial Intelligence
For decades, chess has been one of the most important laboratories for artificial intelligence. Drawing on personal experience with several generations of chess computers, this session traces the development from human–machine competition to cooperation with superhuman systems. It asks what chess can teach us about trusting AI, retaining meaningful human agency and making responsible decisions in a world increasingly shaped by intelligent machines.

Literature

• Der Königsplan – Strategien für Ihren Erfolg Kindermann/von Weizsäcker Rowohlt

Lecturer

Intuition trainer and keynote speaker Stefan Kindermann is an international chess grandmaster and co-founder of the Munich Chess Academy and the Munich Chess Foundation . He has been a member of Rotary International Munich since 2012. He is the author of several specialist books, a columnist for the Süddeutsche Zeitung, an NLP Master Practitioner, a keynote speaker, a trainer, and a coach. Together with Prof. Robert von Weizsäcker and Dijana Dengler, he developed the Königsplan strategy model and teaches this concept through lectures, coaching sessions, and seminars. During his time as a professional chess player, he participated in eight Chess Olympiads and one World Championship, and with Bayern Munich he won the German Team Championship nine times as well as the European Cup. Stefan Kindermann currently plays in the first Bundesliga for MSA Zugzwang, the team of the Munich Chess Academy.

Affiliation: Münchener Schachakademie GmbH und Münchener Schachstiftung
Homepage: www.koenigsplan.com

Lecturer: Robert Porzel, Laura Spillner
Fields: Computational Linguistics

Content

1. Session 1 will introduce the branches of natural language processing pertinent to the rise of LLMS
2. Session 2 will introduce early language models and what they were used for
3. Session 3 will introduce common tasks, such as classification, that are handled by ML approaches
4. Session 4 will introduce embeddings and the types of neural networks that paved the way for LLMS

Literature

• Daniel Jurafsky and James H. Martin. 2026. Speech and Language Processing: An Introduction to Natural Language Processing, Computational Linguistics, and Speech Recognition with Language Models, 3rd edition. Online manuscript released August 19, 2026.
• https://web.stanford.edu/~jurafsky/slp3.

Lecturer

Throughout his studies Robert Porzel earned degrees in general linguistics, computational linguistics and computer science. Previous to his current work he worked in the interdisciplinary collaborative research center “Language and Situation” (SFB 245) and the special research program “Language Production” (SPP 1022) funded by the German Research Foundation (DFG). Thereafter, he worked at the European Media Laboratory (EML) on several natural language processing projects, i.e. DeepMap, Embassi, SmartKom and SmartWeb, which were funded by the Klaus Tschira Foundation (KTS) and the German Ministry for Education and Research (BmBF). In between he was a guest researcher at the University of California, Berkeley working on a formalization of embodied construction grammar in the EDU project. After moving to the University of Bremen, Robert Porzel started work on contextual computing for autonomous collaborative systems in the DFG-funded special research program “Autonomous Logistic Processes” (SFB 637). He initiated and chaired the biannual ScaNaLU workshop series on scalable natural language understanding systems (2002, 2004 and 2006) and was a member of the steering board of the Interdisciplinary College (IK) – an international spring school on cognitive science, neurosciences and AI – which he chaired in 2010 together with Natalie Sebanz and Manfred Spitzer. His current position is that of “Reader in Digital Media” at the University of Bremen and he researches in the collaborative research center “Everyday Activity Science and Engineering” (SFB 1320) on knowledge models for cognitive robotics.

Affiliation: University of Bremen
Homepage: https://www.uni-bremen.de/dmlab/team/dr-ing-robert-porzel

Laura Spillner is a doctoral researcher at the Digital Media Lab. She studied media science and digital media and then completed a master’s degree in computer science at the University of Bremen, with a focus on artificial intelligence, cognition and robotics. Her master’s thesis focused on natural language processing and conversational agents. Laura is working in the European Project “MUHAI: Meaning and understanding in Human-centered artificial intelligence”. In this project and in her PhD thesis she is researching natural language understanding, and the development of explainable algorithms AI and hybrid AI in the field of language processing.

Affiliation: Universität Bremen
Homepage: https://www.uni-bremen.de/en/dmlab/team/laura-spillner

Lecturer: Asena Boyadzhieva
Fields: Cognitive Science, Embodied Cognition, Neuroscience, Psychology, Systems Thinking, Social Cognition

Content

How do we remain responsive and capable of acting when the future is uncertain?

This practical course explores resilience as an embodied capacity: the ability to stay present, regulate, connect and act in conditions of complexity and uncertainty. Drawing on embodied cognition, interoception, relational practices and systems thinking, the course investigates how processes that are often studied separately — individual regulation, interpersonal trust, collective dynamics and responsible action — are connected through the living body.

Rather than treating embodiment as an alternative to cognitive or analytical approaches, we use embodied practice as a form of experimental inquiry. Through movement, attention, sensory awareness, relational exercises and reflection, participants will investigate how bodily states shape perception, interaction and the capacity to respond.

The four sessions move from the individual to the collective and from present experience towards future action:
1. Trusting the Self
We begin with interoception, sensory awareness and grounding. Participants explore how attention to bodily states can support orientation and flexibility when certainty is unavailable, and investigate the relationship between regulation, perception and agency.
2. Trusting Others
Trust is explored as an emergent relational process rather than only a cognitive judgement. Through paired and small-group experiments involving movement, boundaries, synchronisation and responsiveness, participants investigate how trust and co-regulation develop between people.
3. Trusting the System – Interbeing
We expand from interpersonal interaction towards collective dynamics. Group-based practices provide an experiential exploration of feedback, synchronisation, differentiation and interdependence, connecting processes within nervous systems to patterns that emerge between people and within larger systems.
4. Trusting the Future
The final session turns towards the future. Building on the previous sessions, participants explore hope not as optimism or certainty, but as a capacity to remain engaged and act under conditions of uncertainty. The course concludes by connecting embodied awareness with agency, responsibility and possible forms of action.

No previous experience with movement, meditation or somatic practices is required. The course is designed as an experiential laboratory in which participants can investigate their own embodied experience while connecting it to questions from cognitive science, social cognition and systems thinking.

Literature

• Varela, F. J., Thompson, E., & Rosch, E. (1991). The Embodied Mind: Cognitive Science and Human Experience. MIT Press.
• Macy, J., & Johnstone, C. (2022). Active Hope: How to Face the Mess We’re in with Unexpected Resilience and Creative Power. New World Library.
• Scharmer, C. O., & Kaufer, K. (2025). Presencing: 7 Practices for Transforming Self, Society, and Business. Berrett-Koehler Publishers.
• Stroh, D. P. (2015). Systems Thinking for Social Change: A Practical Guide to Solving Complex Problems, Avoiding Unintended Consequences, and Achieving Lasting Results. Chelsea Green Publishing.

Lecturer

Asena Boyadzhieva holds a Master’s degree in Cognitive Science from the University of Vienna and a Bachelor’s degree in Biotechnology from IMC Krems. Her path has since moved from studying living systems in the laboratory to exploring them through the lived body. She is a certified social pedagogue, holistic dance facilitator and yoga teacher, and works at the intersection of embodiment, education, and art. Through The HeArt Center and coMetta, she creates experiential spaces where scientific and embodied ways of knowing can meet. Her work is rooted in the curiosity of what happens when we move beyond understanding the body as an object of study and begin to experience it as an active participant in perception, relationship and action. She is particularly interested in embodied resilience, collective processes and how we can cultivate the capacity to remain connected and responsive in times of complexity and uncertainty.

Affiliation: coMetta – Verein für ganzheitliches Lernen und nachhaltige innere Entwicklung
Homepage: theheartcenter.at

Lecturer: Herbert Jaeger
Fields: Physics, neuroscience, AI, computer science, cognitive science, robotics, systems and control, dynamical systems, philosophy, microchip engineering, and more

Content

The current AI revolution is based on gigantic artificial neural networks that suck up gigantic supplies of data and electrical energy. Biological brains are also gigantic neural networks, but they digest only the little information that a single little life delivers, at 20 Watts peak power. Still, with your brain you can do all the things that you can do, – and that no AI robot yet can do – bake cakes and dance and have fun with your friends and ALL THE REST – think about ALL the things you have done in your life. This setting may explain why currently we see a surge of interest (and funding) in unconventional ‘brain-like’ computing technologies. The current buzzword is ‘neuromorphic computing’ (NC). This course introduces you to this field and its fundamental riddles.

Connection to the IK 2027 focus theme: if (big IF) a really, really brain-like computer would one day be built: would it be responsible for you to switch it on? and after you had switched it on, could you trust that this ‘computer’ acts responsibly?

Session 1: When you are thinking of ‘computing’, does your brain ‘compute’? There are two answers to this question, one of which we do understand. The other is the heartbeat of NC research.

Session 2: A bold but helpless attempt to survey the zillion facets and historical sources of NC, plus flashlights on the state of the art.

Sessions 3: All these challenges! technical, physical, mathematical, practical, educational, ethical, philosophical – almost everything-al. Some might be unsolvable.

Session 4: Speciality of the house: formal semantics for NC. What do we mean when we say a brain computes, and what does the brain mean when it does its thing?

(coverage subject to spontaneous change)

Literature

• There are many introductions and surveys, but they are all written from some specific angle. Authors are human (or used to be), and a human being in his/her lifetime can only first see, then survey what fits in their lifespan. But the wider NC fields are so outrageously multi-multi-disciplinary that single authors and, in fact, entire author collectives cannot come anywhere close to a complete overview, even when they try. Two such tries are
• Finocchio, G., Incorvia, J. A. C., … & Bandyopadhyay, S. [altogether > 50 authors] (2024). Roadmap for unconventional computing with nanotechnology. Nano Futures, 8(1), 012001. Open access at https://iopscience.iop.org/article/10.1088/2399-1984/ad299a (written from a non-digital hardware angle)
• Jaeger, H. (2021). Towards a generalized theory comprising digital, neuromorphic and unconventional computing. Neuromorphic Computing and Engineering, 1(1), 012002. Open access at https://iopscience.iop.org/article/10.1088/2634-4386/abf151 (written from a formal theory-building angle)

Lecturer

Herbert Jaeger studied mathematics and psychology in Freiburg (Germany), got his PhD Computer Science / AI in Bielefeld (Germany) and then did a postdoc at the (then) German National Research Institute for Mathematics and Computer Science (GMD) in Sankt Augustin (Germany), where he subsequently founded the research unit ‘Modeling Intelligent Dynamical Systems’ (MINDS); then from 2001 to 2019 he served as professor in the CS department of Jacobs University Bremen (Germany). Since 2019 he has been Professor for Computing in Cognitive Materials at the University of Groningen. Current research focus: mathematical foundations for a theory of computing on the basis of non-digital physical substrates. Jaeger retired in June 2025 and now has almost enough time for tackling the math-of-complex-systems riddles that got him hooked since student times.

Affiliation: University of Groningen
Homepage: https://www.ai.rug.nl/minds/

Lecturer: Benjamin Paaßen
Fields: Machine Learning

Content

Machine learning is concerned with automatically learning models (patterns, regularities, correlations) from training data such that these models generalize to new data. To do so, machine learning combines concepts from mathematics (esp. statistics, probability theory, linear algebra, and optimization), artificial intelligence, and computer science. This course will provide an introduction to machine learning for the un-initiated. Students will need to suffer through some math, but hopefully my enthusiasm will convey the beauty behind it 🙂 And I will employ ample examples and pictures.

In more detail, the course will have four sessions with the following topics:

1. Basic Concepts: Functions, learning algorithms, optimization, linear regression (as an example of a learning algorithm), regularization, probability theory, machine learning theory, how to design a ML experiment, how to read an ML paper
2. Classic machine learning tasks and methods to solve them: The distance perspective on ML, Regression, Classification, Dimensionality Reduction, Clustering
3. Artificial neural networks and deep learning: Neural network modules, recipes for neural networks, including current LLMs, adversarial attacks
4. Reinforcement learning and fairness

Each session is accompanied by a (voluntary) programming exercise in Python. Exercise sheets (and slides) can be found here: https://bpaassen.gitlab.io/Teaching.html

Literature

• Literature is optional and more regarded as \’further/complementary reading\’:
• Barber, D. (2012). Bayesian Reasoning and Machine Learning. Cambridge University Press. Cambridge, UK. http://www.cs.ucl.ac.uk/staff/d.barber/brml/
• Biehl, M. (2023). The Shallow and the Deep: A biased introduction to neural networks and old school machine learning. University of Groningen Press. Groningen, Netherlands. https://doi.org/10.21827/648c59c1a467e
• Goodfellow, I., Bengio, Y., and Courville, A. (2016). Deep Learning. MIT Press. Cambridge, MA, USA. https://www.deeplearningbook.org/
• Paaßen, B. (2019). Lecture Notes on Optimization. Bielefeld University. https://pub.uni-bielefeld.de/record/2935200

Lecturer

Benjamin Paaßen is Junior Professor for Knowledge Representation and Machine Learning at Bielefeld University, Germany. Their focus is on domain-informed, interpretable, and explainable machine learning, with a particular focus on machine learning for education. They are also affiliate researcher at the Educational Technology Lab of the German Research Center for Artificial Intelligence (DFKI), member of the Young College of the Northrhine-Westphalian Academy of Sciences and Arts, and Junior Fellow of the German Computer Science Society.

Affiliation: Bielefeld University
Homepage: https://bpaassen.gitlab.io/

Lecturer: Thomas Eßmeyer
Fields: Human Computer Interaction, Psychology

Content

The design of online content is increasingly governing and disrupting our choices, while the truthfulness of content becomes harder to assess. A root lies in the design of user interfaces, which are often driven by commercial incentives conflicting with users’ agency and their best interests. In a time of generative AI and LLMs, where the design process is frequently offloaded to automated systems, it is all the more important to understand the ethical caveats of modern technologies before they leave our labs. This course will discuss how deceptive design and dark patterns shape our behaviour and expectations, leading to consequences from frustrations to actual harm. We will catch up with the current state of the art behind this research domain, take an excursion to regulatory protective measures, and discuss paths forward to develop human-centred technologies. Below is a preliminary structure for this course, which will be accompanied with interactive elements. This structure might be subject to slight changes:

Session 1 introduces the concept of dark patterns in the context of human-centred design.
Session 2 discusses the cognitive mechanisms and biases at play and risks through generative AI and LLMs.
Session 3 addresses both legal and organisational responsibilities when people are harmed.
Session 4 covers ethical caveats for the development of user interfaces and what we can do better.

Literature

• Colin M. Gray, Cristiana Teixeira Santos, Nataliia Bielova, and Thomas Mildner. 2024. An Ontology of Dark Patterns Knowledge: Foundations, Definitions, and a Pathway for Shared Knowledge-Building. In Proceedings of the CHI Conference on Human Factors in Computing Systems, 1–22. https://doi.org/10.1145/3613904.3642436
• Ana Caraban, Evangelos Karapanos, Daniel Gonçalves, and Pedro Campos. 2019. 23 Ways to Nudge: A Review of Technology-Mediated Nudging in Human-Computer Interaction. In Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems, 1–15. https://doi.org/10.1145/3290605.3300733
• Richard H. Thaler and Cass R. Sunstein. 2008. Nudge: improving decisions about health, wealth, and happiness. Yale University Press, New Haven.
• Arunesh Mathur, Jonathan Mayer, and Mihir Kshirsagar. 2021. What Makes a Dark Pattern … Dark ? Design Attributes, Normative Considerations, and Measurement Methods. In CHI’21, 18. https://doi.org/10.1145/3411764.3445610

Lecturer

After successfully completing a BA in Digital Media and an MSc in Computer Science, Dr Thomas Eßmeyer (né Mildner) received a PhD at the University of Bremen. In his work, Thomas focuses on user wellbeing and countermeasures to unfair and deceptive design practices, often referred to as Dark Patterns.

Affiliation: University of Bremen
Homepage: https://thomasessmeyer.com/

Lecturer: Thorsten O. Zander
Fields: Artificial Intelligence, Neuroscience

Content

Brain–computer interfaces (BCIs) extend the interaction space by translating neurophysiological activity into machine-relevant information. In this course, we treat BCIs as human–computer interfaces that add a direct information channel from the brain, rather than as “mind reading.” We structure the field by interaction function and intent: (i) active/reactive BCIs for direct control and communication, and (ii) passive BCIs that infer covert user state (e.g., workload, attention, error processing, surprise, affect-related responses) to enable neuroadaptive technology, that is, systems that adapt their behavior based on implicit neurophysiological evidence.

We build a coherent end-to-end view of the BCI pipeline: selecting target signals with neuroscientific motivation, acquiring EEG, establishing a design that supports inference, transforming signals into robust features, training and validating models, and integrating outputs into real-time applications. Throughout, we emphasize what typically breaks when moving from clean laboratory calibration to interactive, non-stationary, artifact-rich contexts. In particular, we contrast lab performance metrics with application-oriented evaluation, and we discuss typical failure modes such as artifact learning, overfitting to calibration, and hidden confounds.

The course uses EEG as the main modality because it remains the most widely deployed non-invasive option. We cover what EEG can and cannot represent, why context and task structure shape interpretation, and how time-, frequency-, and time–frequency-based perspectives map to neural rhythms and event-related responses.

Learning outcomes

After the course, participants will be able to:
– Distinguish active/reactive and passive BCI interaction modes and map them to HCI use cases.
– Explain the EEG measurement chain and the major determinants of signal quality and interpretability.
– Design BCI calibration and evaluation paradigms that reduce confounds and support generalization.
– Implement the core analysis logic from raw EEG to features and classification/regression outputs.
– Critically assess validation, identify common pitfalls, and choose evaluation strategies appropriate for real-world deployment.

Sessions

Session 1 — What is a BCI, and what is it for?

We define BCIs as systems that translate brain activity into machine-relevant information and situate them in human–computer interaction. We introduce a functional taxonomy: direct-control BCIs (active/reactive) versus passive BCIs for covert state assessment in neuroadaptive systems. Building on this, we discuss the practical consequences of design choices: user learning versus machine learning, calibration requirements, and the basic operating loop (calibration → model training → online inference → feedback/adaptation). We also introduce the notion of augmenting the information space of an interactive system by adding channels that represent user state and context.

Session 2 — EEG as a measurement and inference substrate

We cover EEG fundamentals needed for BCI work: sensors and impedances, referencing, reproducible placement (10–20/10–10), sampling, filtering, and physiological constraints on scalp-level observability. We connect measurement to experimental design: synchronous versus asynchronous paradigms, causal versus non-causal processing constraints, and how artifact structure (EOG/EMG/motion) can dominate apparent “BCI performance” if not controlled. We discuss how to design calibration tasks that produce valid labels for the target state, and how to reduce confounds using behavioral and peripheral measures.

Session 3 — From EEG to features and models

We treat feature extraction as “setting a focus” on signal aspects relevant to the cognitive or affective target construct. We cover epoching/segmentation, time-domain and spectral representations, spatial filtering, and how features become machine-learning inputs. We then introduce classification/regression logic with an emphasis on separability, class imbalance, and generalization beyond calibration. We discuss why continuous state estimates can be more robust than hard thresholds in non-stationary contexts, and what that implies for system design and validation.

Session 4 — Validation, failure modes, and neuroadaptive HCI case studies

We discuss how to verify whether a BCI meets the assumptions behind its design. We contrast (i) effect-driven validation (performance across calibration/test/application), (ii) data-driven diagnostics (quality indices, artifact sensitivity, distribution shift), and (iii) neuroscience-informed checks (plausible spatio-temporal patterns, condition contrasts). We conclude with passive BCI and neuroadaptive HCI case studies, including error-related and workload/attention-related signals, and we highlight boundary conditions for transfer from lab paradigms to interactive real-world use.

Literature

• Zander, T. O. (2012). Utilizing brain-computer interfaces for human-machine systems. Doctoral dissertation, Technische Universität Berlin. Available at: https://depositonce.tu-berlin.de/bitstreams/d47c3b2f-b2f2-4cd3-b30a-7a7adcf953b9/download
• Vidal, J. J. (1973). Toward direct brain-computer communication. Annual Review of Biophysics and Bioengineering.
• Wolpaw, J. R., Birbaumer, N., McFarland, D. J., Pfurtscheller, G., & Vaughan, T. M. (2002). Brain–computer interfaces for communication and control. Clinical Neurophysiology.
• Wolpaw, J. R., & Wolpaw, E. W. (2012). Brain–Computer Interfaces: Principles and Practice.
• Zander, T. O., & Kothe, C. (2011). Towards passive brain–computer interfaces: applying BCI technology to human–machine systems in general. Journal of Neural Engineering.
• Zander, T. O., & colleagues: Neuroadaptive technology enabling implicit interaction (cursor control). PNAS (2016).
• Brouwer, A.-M., Zander, T. O., et al. (2015). Six recommendations to avoid common pitfalls when using neurophysiological signals for cognitive/affective state inference. Frontiers in Neuroscience.
• Zander, T. O., & Jatzev, S. Context-aware BCIs and the information space of user, technical system, and environment. Journal of Neural Engineering (early 2010s).
• Blankertz, B., et al. (2007). The non-invasive Berlin brain–computer interface. NeuroImage.
• Stern, J. (2013). Atlas of EEG Patterns (2nd ed.).
• Muthukumaraswamy, S. D., Johnson, B. W., & McNair, N. A. (2004). Mu rhythm modulation… Cognitive Brain Research.
• Maeder, C. L., et al. (2012). Pre-stimulus sensorimotor rhythms influence BCI

Lecturer

Prof. Dr. rer. nat. Thorsten O. Zander is Lichtenberg Professor for Neuroadaptive Human–Technology Interaction at Brandenburg University of Technology (BTU) Cottbus-Senftenberg, where he researches neuroadaptive interaction using passive brain-computer interfaces, including implicit interaction with technology, cognitive exploration for the automated guidance of artificial intelligences, and the ethics of neuroadaptive technologies; he previously held postdoctoral positions at TU Berlin (Biological Psychology and Neuroergonomics) and the Max Planck Institute for Intelligent Systems in Tübingen and served as a group leader at TU Berlin. He studied mathematics (with a focus on mathematical logic) at the University of Münster and earned his PhD at TU Berlin on the overarching topic of applying brain-computer interfaces to human-machine systems. His work has been recognized with awards, including the Raja Parasuraman Award (Best Senior Researcher, Neuroergonomics Society) and Best Dissertation of the Willumeit Foundation.

Affiliation: BTU Cottbus-Senftenberg
Homepage: https://www.b-tu.de/fg-neuroadaptive-hci

Lecturer: Ivana Kajić, Philipp Wicke
Fields: Cognitive Science, Linguistics, Artificial Intelligence

Content

This course examines the relationship between language, perception, and intelligence, using recent developments in generative AI as a central case study. Moving from cognitive linguistics to multimodal machine learning systems, the course investigates how systems transition from text-based representations to models that increasingly integrate perception and action. Across four lectures, we move from theoretical foundations to technical architectures and finally to societal and industrial implications.

• Lecture 1 introduces the conceptual foundations of the course. We explore the hypothesis that human thinking is deeply structured by language, examining linguistic universals, linguistic relativity, and the role of metaphor and conceptual framing. Language is presented not merely as a communicative tool but as a generative system that structures world models. This session establishes the idea that if human cognition is scaffolded by language, then language-trained AI systems offer a particularly revealing lens through which to rethink intelligence.
• Lecture 2 shifts the focus to embodiment and perceptual grounding. We examine theories of embodied cognition and consider how bodily experience shapes conceptual systems. The lecture discusses how abstract thought is rooted in sensorimotor experience and presents language as an interface between pre-linguistic cognition and articulated reasoning. By contrasting embodied human cognition with predominantly text-trained AI systems, this session sharpens the central question of the course: can intelligence emerge from language alone, or does meaningful understanding require grounding in perception and action?
• Lecture 3 explores the technical foundations of modern generative AI, moving from large language models (LLMs) to multimodal architectures. After reviewing the core principles of transformer-based language models, the lecture expands to vision–language models, multimodal training paradigms, and large-scale deployment techniques such as retrieval-augmented generation and in-context learning. The session highlights how these systems are developed in practice, the role of human data and alignment, and current challenges including interpretability and safety. By examining how AI systems increasingly integrate text and perception, we assess both their capabilities and structural limitations.
• Lecture 4 turns to real-world applications and broader impact. Rather than focusing exclusively on speculative AGI narratives, this session highlights how AI is already shaping scientific research, industrial processes, and economic infrastructures. We examine examples from scientific discovery, energy optimization, manufacturing, and operations research, alongside ongoing debates around trust, labor, and human–AI interaction. Designed as an interactive and discussion-based session, this lecture also critically evaluates the gap between technological hype and practical implementation, offering a forward-looking yet grounded perspective on the future of multimodal and agentic systems.

Literature

• Kajić, Ivana, et al. “Evaluating numerical reasoning in text-to-image models.” Advances in Neural Information Processing Systems 37 (2024): 42211-42224.
• Kajić, Ivana, and Aida Nematzadeh. “Evaluating Visual Number Discrimination in Deep Neural Networks.” Proceedings of the Annual Meeting of the Cognitive Science Society. Vol. 45. No. 45. 2023.
• Albuquerque, I., Ktena, I., Wiles, O., Kajić, I., Rannen-Triki, A., Vasconcelos, C., & Nematzadeh, A. (2025). Benchmarking Diversity in Image Generation via Attribute-Conditional Human Evaluation. arXiv preprint arXiv:2511.10547.
• Evans, Vyvyan, and Melanie Green. Cognitive linguistics: An introduction. Routledge, 2018.
• Wei, Jason, et al. “Emergent abilities of large language models.” arXiv preprint arXiv:2206.07682 (2022).
• Boroditsky, Lera. “Does language shape thought?: Mandarin and English speakers’ conceptions of time.” Cognitive psychology 43.1 (2001): 1-22.
• Wicke, Philipp, Wachowiak, Lennart. “Exploring Spatial Schema Intuitions in Large Language and Vision Models” ACL 2024 Findings. 
• Wicke, Philipp, and Marianna Bolognesi. “Emoji-based semantic representations for abstract and concrete concepts.” Cognitive processing 21.4 (2020): 615-635.

Lecturer

Ivana Kajić is a Senior Research Scientist at Google DeepMind in Montréal, Canada. Her research interests include applying methods and techniques from cognitive science in analysis and characterization of behavior of machine learning models. Specifically, this includes designing evaluation protocols, benchmarks and metrics to comprehensively understand capabilities and limitations of large vision-language models that in recent years have demonstrated strong performance in a variety of tasks. She completed her PhD thesis titled “Computational Mechanisms of Language Understanding and Use in the Brain and Behaviour” in 2020 at the University of Waterloo in Canada.

Affiliation: Google DeepMind
Homepage: www.ivanakajic.me

Affiliation: BTU Cottbus, AURYAL GmbH
Homepage: www.phil-wicke.com