¥ You do not model the baseline condition
¥ e.g., ÒrestÓ, visual fixation,
high-low tone discrimination, or some
other simple task
¥ FMRI can only measure changes
in MR signal levels between tasks
¥ So you need some simple-ish task to serve as a reference point
¥ The baseline model (e.g., a + b ×t ) takes care of the signal level to which the
MR signal returns when the ÒactiveÓ tasks are turned off
¥ Modeling the reference task explicitly would be redundant (or ÒcollinearÓ, to anticipate a forthcoming jargon word)