|M.Sc Student||Carmeli Tomer|
|Subject||Using Phase Relation Analysis for Finding Physiological|
Correlates of Functional Connectivity during
|Department||Department of Biomedical Engineering||Supervisor||Professor Emeritus Hillel Pratt|
One of the currently most challenging research topics in the field of neuroscience is the one of functional connectivity. In order to perform even the most basic cognitive task, information must be exchanged and integrated between different functional areas supporting sensory processing, cognitive functions, motor systems and so on.
The mechanisms that underlie such large scale neural integrations are still under debate, with several potential candidates. A widely acceptable marker for the manifestation of functional connectivity is the emergence of synchrony in the firing patterns between cell assemblies, typically within the gamma band (25-70 Hertz).
Studying the synchrony between cells assemblies have been mostly conducted using electromagnetic imaging methods, recording single unit spikes, local field potentials or scalp MEG/EEG activity. Further time-frequency analysis is then applied to study reciprocal markers of synchrony such as coherence and phase relations.
In this study, we conducted an Event Related Potential (ERP) experiment with 13 human subjects. Our first purpose was to investigate the effects of the type of information retrieved, on the spatio-temporal distribution of activity during a memory retrieval task. We used standard ERP processing methods to study the differences in the scalp potentials and estimated the intracranial current density distribution. Our second purpose was to investigate the functional connectivity between regions supporting the task of memory retrieval, and how is it affected by the type of information retrieved. We used the estimated current density per voxel within the brain volume, yielding a signal that describes the activity change over time. We then applied band-pass filter and Hilbert transform to extract the instantaneous properties of phase, over a narrow frequency band. Estimation of synchrony between voxels was established using phase-locking statistics. The processing is accompanied by non-parametric statistics in order to establish the significance of the findings.
We have found several regions that were specifically activated at memory retrieval task and some regions whose activation was task dependant. The functional connectivity analysis has been successful in identifying connections specific to memory retrieval.
The procedure outlined in this study is generic in nature and can be applied to data obtained from any ERP source estimation tool, in any ERP task. Furthermore, it could be applied to data obtained by intracranial recordings.