360 h

2 SWS Lecture
Total workload 60h: 30h contact/30h individual revision of lecture contents, test preparation

1 SWS Seminar
Total workload 45h: 15h contact/30h individual reading and test preparation

10.5 SWS Supervised exercise
Total workload 255h: 145h contact/110h individual work on portfolio tasks (programming, interpretation of simulation results)

++ Neurosci. knowlg.
+ Scient. Literature
+ Social skills
++ Interdiscipl. knowlg
++ Maths/Stats/Progr.
+ Data present./disc.

+ Scientific EnglishUpon successful completion of this course, students
• are able to implement and apply algorithms in Matlab
• have learned to handle scientific data independently
• have acquired theoretical and practical knowledge of advanced data analyis techniques
• know about computational model approaches on different levels of abstraction
• know how to perform model simulations for single cells and small neuronal networks
• can interpret simulation results in a neuroscientific context
 

This course consists of six weeks with different topics, which are introduced in lectures, discussed in depth using selected literature in the seminar and consolidated in computer-based hands-on exercises (in Matlab). Portfolio tasks, mainly interpretation of programming results are given every day.

Weeks 1 and 2: Spike train analysis
response tuning, spike triggered average, receptive fields, linear-nonlinear model, spike correlation, linear reconstruction, classification

Weeks 3 and 4: Neuron models
Conductance-based single cell models using differerential equations (passive membrane equation, integrate and fire, Hodgkin Huxley, alpha synapses)

Weeks 5 and 6: Small network models
Feed-forward and feed-back networks, lateral inhibition, central pattern generator, spike-timing dependent plasticity, multi-compartment models

Skripts for each course day will be provided prior to / during the course

Copies of scientific articles for the seminar and as basis for portfolio assignments will be provided prior to the course

Recommended textbooks or other literature:
Dayan / Abbott: Theoretical Neuroscience: Computational and Mathematical Modeling of Neural
Systems. MIT Press (More text book chapters will be suggested prior to the course).

Registration procedure / selection criteria: StudIP; sequence of registration, attandance in pre-meeting

Recommended in combination with:
neu770 Neuroscientific data analysis in Matlab (prior to the course)
neu250 Computational Neuroscience - Statistical Learning (after the course)

during the course

Portfolio, consisting of daily short tests, programming exercises, short reports