Spike Train Statistics from Empirical Facts to Theory: The Case of the Retina - Modeling in Computational Biology and Biomedicine - page 288

head and eyes. As a consequence the captured images used in experiments need
to be dynamically displayed to the retina, reproducing natural motions.

￿ Sophisticated experiment require MEA devices recording from a large numbers

of cells (e.g., > 100). For example a new MEA technology using 4,096 electrodes
matrix makes it possible to recording most of the neurons in a single small
surface, but the recording data string would take several gigabytes of saving space
and requires fast technologies to access data. Moreover, their numerical analysis,
spike sorting + spike train analysis, in any computer or even a cluster will take a
very long time. Actually,

￿ The MEA outputs need adequate spike sorting algorithms able to deal with

larger and larger numbers of cells. Current algorithms allow to treat about 200
electrodes signals.

￿ Spike train analysis necessitates adequate statistical tools applying to a large

number of interacting cells to evaluate different possible models for neural
encoding (e.g., population coding). Current algorithms allow to treat less than
100 neurons for an Ising model.

￿ A validation of any neural coding model is required. This is done by contrasting

its performance against real behavioral sensory results for the animals under
study [ 26 ]. Additionally, without the precise quantification of the animal per-
formance for a particular behavioral task, responding to natural stimulus, it will
not be possible to access the extended validity of any proposed model. Thus, both
the animal capacity and the theoretical model need to be contrasted.

Clearly, those constraints constitute high level challenges for the scientific
community. Probably, this is only the beginning of a long story.

8.7 Online Resources

8.7.1 Database

Webvision. The Organization of the Retina and Visual System
http://webvision.med.utah.edu/
This site summarizes recent advances in knowledge and understanding of the visual
system through dedicated chapters and evolving discussion to serve as a clearing
house for all things related to retina and vision science.

The brain from top to bottom: the retina
http://thebrain.mcgill.ca/flash/d/d 02/d 02 cl/d 02 cl vis/d 02 cl vis.html
This web site contains a series of topics dealing with the brain: “memory and the
brain”, “evolution and the brain” and so on. Each topic is developed at different
levels: beginner, intermediate, advanced. This is a very useful and didactic reference.