|
Homepage of the BMBF funded Bernstein Group
at the Ruhr-Universität Bochum, starting Spring 2007.
Summary
To understand how neural function emerges from the structure of
nervous systems and their sensory and motor surfaces, theoretical
ideas must squarely confront three fundamental challenges. First,
neural function emerges from spatio-temporally continuous processes.
There are virtually no behavioral signatures of the discrete sampling
of representations by individual neurons nor of the temporal
discreteness of individual spikes. Second, strong neuronal
interaction plays a central role in endowing neural systems with
cognitive competences such as decision making and working memory.
Third, neural systems configure for tasks, adapt to ongoing experience
and learn on all time scales, so that inferences about neural function
are quite specific to the stimulus and behavioral context in which
observations are made.
We aim to create a theoretical framework within which these challenges
will be met. Synaptic neural activity will be observed through
real-time optical imaging techniques and parallel electrical
recording. These data lend themselves to the identification of the
underlying network dynamics, which we will achieve using machine
learning techniques. Constructing Distributions of Population
Activation over functionally relevant feature dimensions, we will
estimate neuronal interactions in terms of the width and strength of
interaction kernels. We will investigate theoretically and
experimentally how the representation of visual space supports the
integration of multiple visual features. When perceptual decisions
bring visual objects into the foreground and select associated feature
values, the interactive neural dynamics go through instabilities that
we will characterize theoretically and strive to detect
experimentally. The processes of learning and adaptation will be
studied in theoretical modelling with the longer-term goal of setting
up an approach to their experimental identification.
Our Bernstein Group will thus aim to lay the foundations of a
long-term research program in which theory and experiment are combined
to systematically uncover how higher neural function, including
behavior and elementary forms of cognition, emerge from
spatio-temporally continuous neural dynamics that are closely coupled
to the sensory and motor surfaces and adaptive to sensory and motor
experience.
Principal Invesigators
Members
| Agnieszka Grabska-Barwinska
| 
|
| Valentin Markounikau
|
|
| Nora Nortmann
|
|
| Jan-Hendrik Reimann
|
|
| Sascha Rekauzke
|
|
| Sebastian Schneegans
|
|
Contact
Prof. Gregor Schöner
Institut für Neuroinformatik
Ruhr-Universität Bochum
44780 Bochum
Germany
tel: +49-234-322-7965
fax: +49-234-321-4209
info@computational-neuroscience-bochum.de
|
![[EMail]](images/letter.gif){kind=small}
