Dear Eric,
The manual says "timeseries" not t-stats, which isn't that clear at the second-level. The values essentially refer to the input values for either level - at first-level this is the timeseries (values from the MRI scans after initial pre-stats filtering) while at the second-level they refer to the input of the GLM which are the COPE values (like the beta values in the GLM) as it is these that are passed up for the second-level GLM. We do not analyse the t-stats from the first-level directly, but instead pass up both the COPE and VARCOPE values. It is the first-level COPE values that you are seeing here.
All the best,
Mark
On 10 Jul 2012, at 07:21, Eric Walden wrote:
> Thank you for the reply Steve. However, I am still confused, the manual says, "each of these is also a link to a raw text file giving the data timeseries plot at the position of the maximum image value within the mask" However, that is what is confusing. This could not possibly be the data time series for the t-stats. There are no t-stats at 666.076. I would believe numbers like 2 or 5 or –3, but not 600+. These look a lot more like the actual activation values, but that does not make sense because it is a second level analysis. If it were an activation level, then what time would these be at. Clearly, they are at the peak voxel, but is it the first scan of the experiment, the last scan, the highest activation value of a scan.
>
|