[Erppcatoolkit-support] decomposed individual ERP waves do not resemble to the original inidudual ERP waves?

[Hongying Z. <hy...@ee...>]

Dear ERP PCA tookkit users,

I'm confused about the output, when I applied the toolkit (temporal
PCA) to the sample dataset (introduced in Dien, J., & Frishkoff, G. A.
(2005). Principal components analysis of ERP data. In T. C. Handy
(Ed.), Event related potentials: A methods handbook (pp. 189-208).
Cambridge,Massachusetts; London,England: The MIT Press.)

Since the first 2 factors explained the 100% variance of dataset,2
factors are retained only. After the analysis, I obtained decomposed
factor waves for each subject (the "facWaves" output from PCAtoEGIS
function). But the wave pattern for decomposed factor waves are
different from original individual waves. (Although the grand average
pattern is the same for these two conditions).  Is it reasonable, why?

Here I attached the matlab file for generating the figure (figure 6.
for original individual ERP waves and figure 7. for decomposed
individual factor waves).

Best Regards,
H.Y.Zheng

-- 
==========================================
Ms. Zheng Hongying
Ph.D (Post-candidacy)
Room 310, Ho Sin-Hang Engineering Building
Department of Electronic Engineering
The Chinese University of Hong Kong
Shatin, Hong Kong, China
Tel: +852 2609 8250
Email: hy...@ee...
Lab: www.ee.cuhk.edu.hk/~lel/
==========================================