A toolbox for residue iteration decomposition (RIDE)—A method for the decomposition, reconstruction, and single trial analysis of event related potentials

Talk :      February 19, 12:30 pm in the Karp Conference Room, #501 of the Goodman Cancer Centre, McGill

Workshop:    February 19, 14:05 pm in #208/09 of the McIntyre Medical Centre, McGill

Registration for Workshop : http://www.crblm.ca/events/workshop_using_residue_iteration_decomposition_ride_toolbox_erp_data

Event-related brain potential (ERP) is a powerful and popular tool in cognitive neuroscience, based on the idea that consecutive ERP components are associated with different and specific cognitive subprocesses, providing insight into online processing of events at high time resolution. The traditional averaging method to extract ERPs from the continuous EEG assumes a fixed temporal relationship between the eliciting event and the different ERP components. However, this assumption may be substantially violated in many cases, i.e., the timing of sequential processes vary from trial to trial. Condition or population differences in temporal component variability (jitter) may induce or camouflage differences in the amplitudes of ERP components. Different components within the ERP may be affected to different degrees by jitter-related distortions. Therefore the conventional average ERP is a mixing of blurred components. Yet, there has been no theoretically satisfying and practically applicable solution for this problem. Guang Ouyang and colleagues have recently developed a new method, RIDE (Residue Iteration Decomposition), which separates the ERP into different component clusters with different latency variability. RIDE provides a tool to (1) separate overlapping components, (2) obtain information about amplitudes and latencies of each component on a single trial basis, and (3) reconstruct ERP by compensating the latency variability. In the seminar, he will give an introduction into the algorithms and principles and applications of RIDE. He the following workshop, he will demonstrate the implementation of RIDE.

Registration is required for the workshop (see link above to access the registration page)

Bio:

Guang Ouyang is a postdoctoral fellow in the Department of Physics at Hong Kong Baptist University, Kowloon. Development of the RIDE toolbox was completed as part of his PhD thesis. 

The workshop and seminar are organized and funded by CRBLM ( http://www.crblm.ca ) and Dr. Karsten Steinhauer's Neurocognition of Language Laboratory ( http://www.mcgill.ca/neurocoglab/ )