Hi - this all sounds ok. For practical numerical reasons, it's not trivial to choose a threshold for "rank deficiency" that is appropriate for all cases - as long as your value reported by the warning is greater than say 10e-5 then you can probably happily ignore it. Cheers. On Mon, 5 Jul 2004, Steven Coen wrote: > Hi, > > Using FEAT, I have 3 EV's I want to model - Stimulus A (1 second duration) > Stimulus B (1 second duration) and stimulus C (5 seconds duration) > > Stimuli A or B (randomised stimuli presentation, 20 stimuli each) is > presented every 30 seconds beginning at timepoint zero i.e 0, 30 , 60. > > The third stimuli "C" is subject responses that I am not interested in but > want to put in as a co-variate. "C" occurs every 30 seconds (40 in total) > starting at 15 seconds: ie 15, 45, 75. > > When I put in these three conditions as EV's I get the rank deficiency > message. However, when I code A and B as the same EV i.e simply one > stimulus every 30 seconds for the duration of the experiment and still > include "C" there isn't a problem. I don't understand why!? There are the > same number of stimuli with the same ISI and duration, just coded together. > Can someone explain? > > Many Thanks for your help, > > Steve. > Stephen M. Smith DPhil Associate Director, FMRIB and Analysis Research Coordinator Oxford University Centre for Functional MRI of the Brain John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve