Email discussion lists for the UK Education and Research communities

## SPM@JISCMAIL.AC.UK

#### View:

 Message: [ First | Previous | Next | Last ] By Topic: [ First | Previous | Next | Last ] By Author: [ First | Previous | Next | Last ] Font: Proportional Font

#### Options

Subject:

parameters for modelling

From:

Date:

Fri, 4 Nov 2011 00:07:14 +0000

Content-Type:

text/plain

Parts/Attachments:

 text/plain (24 lines)
 ```I'm new to SPM & fMRI. I've started by trying to modify some existing code to use on my data. I have searched in the archives, but I wasn't able to understand the answers enough. Are there any documents or websites that answer detailed questions like I have? In the code below: - How do you decide what to set T & T0 to when setting up the parameters for your model? Is T (# of time bins per scan) chosen arbitrarily? - Is T0 referring to the slice # or the time? - Is this formula correct: T0 = (TA/2)/TR * T ? (where TA is the duration of the scan) - How is length chosen? (what is window length?)         SPM.xBF.T = 16; % number of time bins per scan         SPM.xBF.T0 = 1; % first time bin (see slice timing) - middle of TA         SPM.xBF.UNITS = 'scans'; % OPTIONS: 'scans'|'secs' for onsets         SPM.xBF.Volterra = 1; % OPTIONS: 1|2 = order of convolution         SPM.xBF.name = 'hrf';         SPM.xBF.length = 32; % length in seconds         SPM.xBF.order = 1; % order of basis set I'm using sparse sampling, and I've read some suggestions to use FIR instead of HRF to model that, but the choice isn't clear. Also, for sparse sampling, I'm still not clear on what to use as the TR: in my study, there is a 8.5 second auditory stimulus with no scanning, followed by a 2 second scan - so is my TR (for the purposes of SPM modelling) 10.5 or 2? Thanks. ```