I agree with Christian that it's better to include a new covariate for
errors. If there aren't many errors and you aren't planning on doing
anything with them (in terms of contrasts), you could probably just create
a single EV and lump all the errors together. Personally I'd stick to
3-column format for all the EVs just because 1-column is a pain unless you
have a specific measurement each TR (e.g. estimated motion parameters).
Joseph Devlin, Ph. D.
FMRIB, Dept. of Clinical Neurology
University of Oxford
John Radcliffe Hospital
Headley Way, Headington
Oxford OX3 9DU, U.K.
Phone: +44 (0)1865 222 494
Fax: +44 (0)1865 222 717
Email: [log in to unmask]