It's not surprising that the amplitude of the model differs for 200ms vs. 1200ms events. This is the basis for the difference in sensitivity between event-related designs (ie short events) vs. blocked-designs (which are equivalent to long events). If you don't expect your BOLD response to exhibit this effect over this particular range, then by all means rescale the amplitude. On the other hand, it is important to note that this is the expected behaviour of the model. Joe -------------------- Joseph T. Devlin, Ph. D. FMRIB Centre, Dept. of Clinical Neurology University of Oxford John Radcliffe Hospital Headley Way, Headington Oxford OX3 9DU Phone: 01865 222 738 Email: [log in to unmask]