BioNB 420: Topics in Systems and Computational Neuroscience
Course Logistics
Thursdays 8:40-9.55 am in B150 Comstock
Christiane Linster
Dept. of Neurobiology and Behavior
W245 Mudd Hall
Ithaca, NY 14853
Phone: 254 4331
email: cl243@cornell.edu
http://cpl.cornell.edu
Office: W249 Mudd Hall
Office hours: Email mailto:cl243@cornell.edu to make appointment
Course webpage: http://www.nbb.cornell.edu/neurobio/linster/BioNB420/BioNB420.htm
Course description
This course will discuss how experimental and computational neuroscience complement each other in systems neuroscience. One focus of the course will be how recent experimental discoveries at the cellular and circuit level have influenced our thinking in the areas of neural coding and learning and memory. A second focus will be how computationally oriented models of neural coding can be applied to results from behavioral experiments. Throughout the course, experimental data as well as computational modeling will be discussed in the framework of systems neuroscience.
This year, the course will cover four topics: (1) Hebbian learning rules; (2) Neural assemblies, synchrony and oscillations; (3) Classical conditioning and the dopamine reward system and (4) Cortical associative memory function and cholinergic modulation. For each topic, we will read and discuss both experimental and modeling papers.
SYLLABUS
Some matlab code will be developed in class.
Course deadlines:
Homework is always due WEDNESDAY at NOON before the thursday lecture.
February 22nd: first version of research paper 1 is due (Hebbian learning)
March 1st : final version of research paper 1 is due.
March 29: first version of research paper 2 (neural assembly and oscillations) is due.
April 5th: final version of research paper 2 is due.
April 12th: first version of research paper 3 (dopamine reward system) is due.
April 19th: final version of research paper 3 is due.
Each student chooses two out of these three topics to write a short paper (see details below).
May 10rd: Matlab simulations are due.
Each student chooses one topic to program a small simulation on (see details below).
Readings: For each lecture, a number of readings will be assigned. In some cases, I will point out which paper should absolutely be read and which need not be read in detail. In case of doubt, read them all or email me! Class participation is a big part of the grade and reading the papers will ensure that you have things to contribute during class.
If you find a paper very hard , DON'T PANIC! Especially the modeling papers that we will read are difficult. Don't worry if you cannot understand everything. Try to get the essentials from the introduction and from the discussion. Email me or come to see me if you want help!
For many papers, I will prepare handouts which will hopefully make the reading easier for you by explaining some of the details that I think are hard to understand or that you cannot possible know. Remember, questions count as class participation, so ask about whatever you did not understand!
For most readings links to the journal are on the webpage, some you need to use medline to download. All of them are accessible through the Cornell library system.
PDFs of notes and handouts I wrote can be accessed directly from the syllabus.
Student presentations: The class will rely heavily on student presentations of certain papers. When it is your turn, read the paper carefully and prepare a presentation of the paper which should take approximately 15 minutes. It is important that your presentation include an introduction of the subject matter, a description of the methods used in the paper, the main results and a short discussion of what you think is relevant in this paper. Be critical! NO POWERPOINT! Use the chalkboard. Remember that everybody in the room has a copy of the paper if you need to refer to figures from the paper.
The schedule for student presentations will be available on the web at: http://www.nbb.cornell.edu/neurobio/linster/BioNB420/presentations.htm
Homework: There will be some homework. Homework will consist of questions about the readings. The homework serves mainly to let me know what you did or did not understand about the readings and gives you an opportunity to ask questions. Homework assignements will be posted on the web by the day of the lecture. You will have to email your completed assigments to me by WEDNESDAY, NOON before the Thursday lecture. Late homework will NOT be considered!.
Papers: Throughout the semester, you will have to write two short research papers. Each paper will cover one of the topics that we will cover in the lectures (hebbian learning, neural assemblies, oscillations and the dopamine reward system). These are five topics out of which you can choose three. (You don't have to write a paper for the last topic in order not to fill up the end of the semester with to much work). These papers should consist of a short (NO MORE THAN 5-10 PAGES) summary of what we discussed in class, and in addition you should discuss what you think of the questions raised, present your criticisms and opinions. It is very important that you try to develop your OWN opinions on these subjects in the papers. For each paper, I will give you comments before you hand in the final version which will be graded. It is important that you read additional literature on the topic when you write these papers.
Simulations. You will write a simple neural simulation in MatLab based on one of the topics that we will discuss in class. I will help with that and provide code wherever necessary. These do not have to be sophisticated.