Hi Joel,

By "no baseline" do you mean that as each task/condition ends, the next one immediately begins? If so then the 8 condition design below may be problematic, one approach here is to treat one condition ( e.g neg-NULL ) as baseline and so remove its EV from the design. The remaining EVs will then effectively model response relative to this baseline - your contrasts ( for the NULL comparison below ) become:

[1 0 0 0 0 0 -1 ]
[0 1 0 0 0 0 0 ]
[0 0 1 0 0 0 -1 ]
[0 0 0 1 0 0 0 ]
[0 0 0 0 1 0 -1 ]
[0 0 0 0 0 1 0 ]

All this can be done at first-level - there is no requirement for a “mid-level” analysis for these contrasts.

Hope this helps,

Kind Regards

Matthew

--------------------------------
Dr Matthew Webster
FMRIB Centre
John Radcliffe Hospital
University of Oxford

On 10 Aug 2022, at 13:16, Joel Patchitt <[log in to unmask]> wrote:

Hi All,

I have data for a study that has a few flaws. First of all the data has no baseline and is all task with no null trials or jitter times.

condition:

1. physiological condition (4) - higher, lower, same, NULL
2. emotion condition (2) - positive, negative

I have created 8 onset files to generate EV's for these conditions:

1 pos-higher 2 neg-higher 3 pos-lower 4 neg-lower 5 pos-same 6 neg-same 7 pos-NULL 8 neg-NULL

I was advised to set up the first level like this:

[1 0 0 0 0 0 0 0]
[0 1 0 0 0 0 0 0]
[0 0 1 0 0 0 0 0]
[0 0 0 1 0 0 0 0]
[0 0 0 0 1 0 0 0]
[0 0 0 0 0 1 0 0]
[0 0 0 0 0 0 1 0]
[0 0 0 0 0 0 0 1]

I know that in SPM you can create complex higher level models based on the outputs of the above first level matrix, but I am not sure how this is done in FSL.

I would like to compare the physiologically present trials against the physiologically NULL trials

Essentially I want to generate this at the second level.

[1 0 0 0 0 0 -1 0]
[0 1 0 0 0 0 0 -1]
[0 0 1 0 0 0 -1 0]
[0 0 0 1 0 0 0 -1]
[0 0 0 0 1 0 -1 0]
[0 0 0 0 0 1 0 -1]

Is this possible in FSL and how would I go about doing it?

Alternatively I could just set up the above contrasts in the first level and run a one sample t-test on each of the 6 COPE images. Seeing as I don't have any baseline I believe this might be the best course of action. What are your thoughts?

Best,
Joel

########################################################################

To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=FSL&A=1

This message was issued to members of www.jiscmail.ac.uk/FSL, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/