Thanks to all who responded to my query a couple weeks ago about correction of ocular artefacts in EEG. The reference that I've found most helpful is
Brunia & al., 'Correcting ocular artifacts in the EEG: a comparison of several methods', Journal of Psychophysiology 3:1-50 (1989).
Some other references - but not a complete list - are present in an old report of mine at http://salk.edu/~matthew/EEG/references.html
I have a frequency-domain correction program working - it will, however, significantly attenuate ERPs. It reads both the Scan386 continuous multiplexed (.CNT) data format and the raw data format of the San Diego EPL system, and it writes Scan386 epoched (.EEG) files. I'm still not so sure about the relative merits of correction in the time and frequency domains, but anyone's welcome to take a look at this program as a guide, for what it's worth. You can access it from my home page (see below). I've compiled it with Microsoft QuickC on an 80486 system but I've written it for portability.
Apropos of the World Wide Web, would anyone be interested in a Web page devoted to ocular correction methods, or perhaps to quantitative EEG in general? If there are prospective contributors I'd be willing to set one up.
BTW, to the person at panix.com who wrote me to say that he was working on similar problems - your message was sent to a different account, and lost during forwarding, so that's why I've been unable to reply.
mkb4 at Cornell.eduhttp://salk.edu/~matthew/