Document Actions

You are here: Home / The Network / Network Partners / Bernstein Collaborations / Memory Network

Memory Network

Bernstein Collaboration for Computational Neuroscience (BCOL)

The main goal of the collaboration is the instantiation of real-time tools for the study of fast dynamical changes in neuronal population activity subserving cognitive processing. It is intended to determine the composition of memory related population activity in monkey prefrontal cortex and manipulate the availability of some of the contributing units during memory maintenance and recall by micro-stimulation. The basic approach for the identification of neuronal assemblies consists of the analysis of patterns of firing rates and higher order neuronal interactions. A recording and analysis tool will be provided to the community, which will make it possible to perform "closed loop" neurophysiological experiments in a behaving animal, based on reversible manipulation of neuronal population activity through spatiotemporally patterned microstimulation. The main tasks of this project comprise the development of a new adaptive spike-sorting algorithm for the detection of single-unit activity, its implementation in a real-time system that allows for driving micro-manipulation devices (automatic positioning and stabilization), the development of a microstimulation device that can be controlled by real-time analysis tools and its application to experiments investigating memory processes in non-human primates.

Subproject 1: Development of a microelectrode array with an automated positioning system and an integrated adaptive fine tuning mechanism for real-time recording of neuronal signals and generation of microstimulation patterns

Subproject 2: Testing and application of real-time assembly identification and manipulation

Subproject 3: Adaptive algorithms for the detection and discrimination of spikes in multi-tetrode recordings (spike sorting)