psy220 - Human Computer Interaction (Vollständige Modulbeschreibung)

psy220 - Human Computer Interaction (Vollständige Modulbeschreibung)

Originalfassung Englisch PDF Download
Modulbezeichnung Human Computer Interaction
Modulkürzel psy220
Kreditpunkte 6.0 KP
Workload 180 h
Einrichtungsverzeichnis Department für Psychologie
Verwendbarkeit des Moduls
  • Master Neurocognitive Psychology (Master) > Mastermodule
Zuständige Personen
  • Rieger, Jochem (Modulverantwortung)
  • Rieger, Jochem (Prüfungsberechtigt)
Teilnahmevoraussetzungen
Enrolment in Master's programme Neurocognitive Psychology or other programs related to the field (e.g. neuroscience, computer science, physics etc.).
Kompetenzziele
Goals of module:
The goal of the module is to provide students with basic skills required to plan, implement and
evaluate brain computer interfaces as devices for human computer interaction. BCIs are an
ideal showcase as they fully span the interdisciplinary field of HCI design, implementation and
evaluation. Importantly, classical BCI-methods can be used for modern data-driven basic neuroscience. 
The module is designed as an "enabler course", meaning that ideally students should be able to understand and start independent studies into the BCI-methods. Therefore, it goes into depth instead of breadth. Good programming skills and some active knowledge of high school maths is strongly advised to maximize the learning outcome.

Competencies:
++ understanding of the foundations of statistical learning techniques
+ basics to understand technical time series processing and machine learning papers
++ interdisciplinary knowledge & thinking
+ experimental methods
++ statistics & scientific programming
+ critical & analytical thinking
+ scientific communication skills
+ knowledge transfer
+ group work
+ project & time management
Modulinhalte
The module will introduce classic BCI paradigms and brain recoding techniques. However the main focus will be on a deeper understanding of the most important signal processing, machine learning, and performance evaluation techniques. The module combines a lecture on the theoretical foundations a seminar/hands on course in which students learn to implement the BCI-processing steps on real neurophysiological data and further elaborate specific subtopics.

Part 1: HCI and BCI Lecture: (Lecture on methodological foundations of BCI): summer

Part 2: Hands on BCI implementation (practical seminar): summer

Topics covered:
  • A brief history of BCIs and examples of HCI control and basic neuroscience using BCI
  • techniques.
  • Data preprocessing (e.g. filtering, projection techniques) and common artifacts and
  • artifact treatment)
  • Feature generation (e.g. fourier transform, spectral estimation techniques, principle
  • components)
  • Machine learning for classification and regression (e.g. model parameter optimization in
  • multivariate regression)
  • Evaluation (e.g. measures of model quality, cross validation to test model generalization,
  • permutation tests)
Where possible the lecture provides mathematical backgrounds of the data analysis techniques.
The practical seminar implements BCI techniques on a real data set and further elaborates
specific topics in seminar form.
Literaturempfehlungen
There is no required textbook. The lecture slides and notes should be sufficient. However some
resources from which they were developed on are given below:

General tutorial text providing and overview and accompanying python code on github:

Holdgraf, Christopher R., Jochem W. Rieger, Cristiano Micheli, Stephanie Martin, Robert T.
Knight, and Frederic E. Theunissen. 2017. “Encoding and Decoding Models in Cognitive
Electrophysiology.” Frontiers in Systems Neuroscience 11.

https://doi.org/10.3389/fnsys.2017.00061. (open access)

Signal processing:

Semmlow, J. L. (2008). Biosignal and medical image processing. CRC press. Basis of most of
the signal processing section. Has some matlab code.

PCA & SVD

Shlens, Jonathon. 2014. “A Tutorial on Principal Component Analysis.” ArXiv:1404.1100 [Cs,
Stat], April.
http://arxiv.org/abs/1404.1100. Great accessible tutorial on PCA

Unsupervised feature Learning and deep learning tutorial:


http://deeplearning.stanford.edu/tutorial/ Basis of the multivariate machine learning techniques.
Has some matlab code.

General texts:

Machine learning and AI:

Hastie, Tibshirani, and Friedman. The elements of statistical learning. Covers a wide range of
machine learning topics. Free online.

Russell and Norvig. Artificial Intelligence: A Modern Approach. A comprehensive reference
BCI

Dornhege et al. (2007) Toward Brain Machine Interfacing, The MIT-Press. A collection of essays
on BCI related topics.

Additional literature and material will be provided on the course website.
Links
Unterrichtssprache Englisch
Dauer in Semestern 1 Semester
Angebotsrhythmus Modul The module will be offered every summer term.
Aufnahmekapazität Modul 15
Hinweise
We strongly recommend to take either psy170, psy270, psy280, or psy220 to gain methodological competencies (EEG, fMRI, TBS, HCI) that are needed for most practical projects and Master's theses!
Modulart Wahlpflicht / Elective
Modullevel MM (Mastermodul / Master module)
Lehr-/Lernform Part 1: lecture; Part 2: practical seminar
Vorkenntnisse Basic programming skills, some high-school level maths
Lehrveranstaltungsform Kommentar SWS Angebotsrhythmus Workload Präsenz
Vorlesung 2 SoSe 28
Seminar 2 SoSe 28
Präsenzzeit Modul insgesamt 56 h
Prüfung Prüfungszeiten Prüfungsform
Gesamtmodul
last lecture in summer term
The module will be evaluated with an oral exam (max. 20 min). 

Required active participation for gaining credits:
1-2 presentations
max. 24 programming exercises in the seminar
participation in discussions on other presentations
attendance of at least 70% in the seminar within one semester (will be checked in StudIP).