Looking at ERPs from Another Perspective: Polynomial Feature Analysis S Straube1, D Feess2 |
|---|
1Robotics Group, University of Bremen, Germany |
| Event-related potentials (ERPs) are classically studied measuring amplitude and latency characteristics of individual components. Such analysis is restricted to individual time points and largely ignores the temporal structure of the ERP. This motivates alternative pre-processing algorithms that might reveal new information about the signal encoded in the temporal relationships between neighbouring data points. In the current work, we fitted polynomials of orders one to four to ERPs (average and individual epochs) before analyzing the signal. Depending on other pre-processing methods (like subsampling and filtering), a low order polynomial should be able to capture the ERP’s shape and reduce noise in single-trials. The polynomials were fitted on local segments over the whole epoch and the subsequent analysis was performed with the resulting coefficients instead of the amplitude values. For evaluation we used data from an oddball task evoking a broad P300 component (five subjects, two sessions each). The descriptive quality of the coefficients was derived from the performance of a support-vector machine classifying ERPs labelled as ‘standard’ and ‘target’, respectively. The corresponding ERP topographies (both, average and single epochs) strengthen the notion that analysis of polynomial features provides a tool for exploration of new relationships in ERP data. |
| Up Home |
|---|