ELEC-E7851 Computational User Interface Design (5 cr, PII)
The course offers a solid introduction to HCI to students who seek principled understanding of user interfaces now and in the future. It introduces the computational and behavioral principles of interface design for students with computational and engineering background. Students learn to formulate design problems and derive solutions by analysis, simulation, and optimization.
The contents cover HCI models broadly, starting from biomechanics and continuing to vision science, psychophysics, cognitive psychology, control theory, and decision sciences. Constructive approaches include optimization, Bayesian methods, and reinforcement learning. The lectures introduce the models and theories, whereas the exercises apply them to realistic problems using pen and paper or code. Application areas cover the most widely used interfaces, including buttons, keyboards, menus, displays, visualizations, input methods, audio and multimedia systems, graphical user interfaces, hypertext etc.
Understanding of computational and model-based approaches to user interface design; Knowledge of major predictive models and theories in human-computer interaction; Ability to formulate and solve realistic user interface design problems using formal and computational approaches.
A non-comprehensive list of examples of competences learned:
- Formal analysis of design problems, including objective functions and design spaces
- Classification methods (e.g., SVM) for input recognition
- Decoders for natural language input
- Combinatorial optimization of assignment problems in layout design
- Bayesian approach to noisy input data, for example in key-target resizing
- Control-theoretical analysis of input transfer functions
- Vision modeling to enhance color perception, saliency, readability, clutter, and discriminability
- Analytical and simulation models to analyze biomechanical requirements of input and fatigue
- Information-theoretical optimization of performance requirements (e.g., in games)
- Cognitive modeling of task performance with visual interfaces (e.g., KLM, CogTool)
- Analysis of multitasking performance in interactive tasks (e.g., MRT, Distract-R)
- State transition analyses to identify sources and consequences of human error
- Computational rationality models of how users adapt (e.g., reinforcement learning)
- Predictive model of stress arousal and appraisal
- Applying perceptual illusions to movement design (e.g., pseudo-haptics)
- Applying psychophysical functions to optimize physical feedback
Prof. Antti Oulasvirta and guest lecturers: Dr. Daryl Weir, Dr. Jussi Jokinen, MSc Kumaripaba Athukorala, MSc Janin Koch, MSc Anna Feit. Teaching assistants: Mika Jokiniemi and Vivek Dhakal.
ELEC-E7850 User Interfaces
Lectures on user interfaces, models, and theories; Assignments solved with Python or Matlab; Assigned readings; Exam
Mandatory programming languages: Python and Matlab. We also recommend a previous course on human-computer interaction. From Aalto University, we recommend CS-C3210 Human-Computer Interaction or ELEC-E7890 User Research.
Note: assignments require familiarity with some general concepts of computational sciences and basic mathematics. Previous courses in behavioral or medical sciences will be beneficial, too.
Grading is based on points earned in assignments and exam:
- Assignments max 100 points; minimum for passing 50
- Exam max 50 points; minimum for passing 25
Due to the type of teaching and materials, participation in lectures is mandatory. A maximum of 4 lectures can be skipped.
The maximum number of students is limited. Students of the University of Helsinki are welcome join the course but are asked to contact the teacher in advance.