Dear Baxter,
> I am learning to use SPM 99 to generate experiments. It seems that the
> onset times for the various stimuli are contained in the matrix
> Sess{1}.ons{n} for the nth event type.
>
> I've noticed with several different designs that these onset times are
> 0.125 seconds later than I would expect: for example, the beginning of
> the first epoch for a block design -- Sess{1}.ons{1}(1) -- is 24.125
> sec when I specify 12 scans (at 2 sec TR) as the time to first trial.
> This also occurs for event-related designs.
This is a reflection of the fact that onset times are specifed in
time-bins of TR/16 = 0.125s for you. One starts acquiring the first
scan at t = 0 seconds (scans) and finishes at t = TR seconds (1 scan).
To ensure indices do not start at 0, 1 time-bin is added to every
onset. 125ms does not matter one way or the other given the time
constants of the hemodyanic response.
I hope this helps - karl
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|