Hi Steve,
Thanks for your suggestion to alter/remove EVs 3 to 6 from my contrasts.
However I'm confused by this suggestion since you indicated in the past that
EVs 3 and 4 seemed reasonable - I refer to my email and your response in
August below (I apologise for not including it in my previous email this
week, I assumed the online JISCMAIL email application would automatically
append it).
So below this email is my initial (August) email and your response regarding
design and contrasts for a paired t-test with positive correlations with
behavioural scores, followed by my recent email (this week) and your
response regarding the addition of 2 more contrasts (to test for negative
correlations). I'd be grateful if you could confirm that the set-up I had in
August was, in hindsight, actually incorrect.
From your suggestion this week, I gather I should be using design and
contrast matrices like this:
Inputs Group EV1 EV2 EV3 EV4 EV5 EV6
SessAsubj1 1 1 1 0 0 0 -2.25
SessAsubj2 1 1 0 1 0 0 -2.25
SessAsubj3 1 1 0 0 1 0 0.75
SessAsubj4 1 1 0 0 0 1 3.75
SessBsubj1 1 -1 1 0 0 0 2.25
SessBsubj2 1 -1 0 1 0 0 2.25
SessBsubj3 1 -1 0 0 1 0 -0.75
SessBsubj4 1 -1 0 0 0 1 -3.75
EV1 EV2 EV3 EV4 EV5 EV6
Con1 A-B [ 1 0 0 0 0 0 ]
Con2 B-A [-1 0 0 0 0 0 ]
Con3 mod+ [ 0 0 0 0 0 1 ]
Con4 mod- [ 0 0 0 0 0 -1 ]
Please note Con3 and Con4 above, as per your recent instruction. While these
contrasts test for positive and negative correlations with EV6 (behavioural
difference score), respectively, I'm having trouble understanding how those
correlations relate to paired differences between sessions, i.e. both A-B
and B-A contrasts, which is what I'm interested in. Specifically, in
addition to the standard paired t-test contrasts, I am interested in a set
of contrasts that will tell me where:
1. A>B and positively correlated with behavioural performance difference
score,
2. B>A and positively correlated with behav perf diff score,
3. A>B and negatively correlated with behav perf diff score, and
4. B>A and negatively correlated with behav perf diff score.
Do Con3 and Con4 above tell me any/some of these things?
I apologise for my tenuous understanding of design and contrast matrices.
Many thanks for your assistance,
Emma
------------------------------------------
Hi, yes, this looks perfect I think.
Cheers.
On 6 Aug 2008, at 04:40, E. Bendall wrote:
> Dear FSLers,
>
> I would like some clarification regarding contrast setup for a paired
> t-test with additional EV for correlation.
>
> I am conducting a paired t-test on FA images for subjects imaged at
> two sessions. I am interested in the correlation of any FA differences
> with differences in a performance score (also collected at session 1
> and session 2).
>
> My design matrix is set up like the paired t-test example in the FEAT
> manual page but with an additional EV containing the differential
> performance scores (demeaned and orthogonalised wrt EV1).
>
> Inputs Group EV1 EV2 EV3 EV4 EV5 EV6
> SessAsubj1 1 1 1 0 0 0 -2.25
> SessAsubj2 1 1 0 1 0 0 -2.25
> SessAsubj3 1 1 0 0 1 0 0.75
> SessAsubj4 1 1 0 0 0 1 3.75
> SessBsubj1 1 -1 1 0 0 0 2.25
> SessBsubj2 1 -1 0 1 0 0 2.25
> SessBsubj3 1 -1 0 0 1 0 -0.75
> SessBsubj4 1 -1 0 0 0 1 -3.75
>
> EV1 EV2 EV3 EV4 EV5 EV6
> Con1 A-B [ 1 0 0 0 0 0 ]
> Con2 B-A [-1 0 0 0 0 0 ]
> Con3 A-Bmod [ 1 0 0 0 0 1 ]
> Con4 B-Amod [-1 0 0 0 0 1 ]
>
>
> Con1 and Con2 are the differential contrasts to test for FA change
> between sessions. Con3 and Con4 are intended to test for FA change
> between sessions that also correlate with a change in performance
> score. Are Con3 and
> Con4
> set up correctly?
>
> Many thanks in advance,
>
> Emma
>
---------------------------------------------------------------------------
Stephen M. Smith, Professor of Biomedical Engineering Associate Director,
Oxford University FMRIB Centre
FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
---------------------------------------------------------------------------
-----Original Message-----
From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf
Of Steve Smith
Sent: Tuesday, 25 November 2008 6:08 PM
To: [log in to unmask]
Subject: Re: [FSL] Paired t-test contrasts - correlation with additional
variable
Hi,
On 24 Nov 2008, at 06:26, Emma Bendall wrote:
> Hi Steve and FSLers,
>
> I have run into a problem that I hope someone can shed light on. I
> refer to
> the design matrix in my previous email (below) with additional
> contrasts to
> allow for testing for negative as well as positive correlations. To
> reiterate, I am using a paired t-test on FA data obtained for a
> group at two
> separate time points with an additional EV (behvioural difference
> scores,
> demeaned and orthogonalised wrt EV1). The FA data was processed
> using TBSS.
> The design matrix is similar to:
>
> Inputs Group EV1 EV2 EV3 EV4 EV5 EV6
> SessA_s1 1 1 1 0 0 0 -2.25
> SessA_s2 1 1 0 1 0 0 -2.25
> SessA_s3 1 1 0 0 1 0 0.75
> SessA_s4 1 1 0 0 0 1 3.75
> SessB_s1 1 -1 1 0 0 0 2.25
> SessB_s2 1 -1 0 1 0 0 2.25
> SessB_s3 1 -1 0 0 1 0 -0.75
> SessB_s4 1 -1 0 0 0 1 -3.75
>
> With contrasts:
> EV1 EV2 EV3 EV4 EV5 EV6
> Con1 A-B [ 1 0 0 0 0 0 ]
> Con2 B-A [ -1 0 0 0 0 0 ]
> Con3 A-Bmod+ve [ 1 0 0 0 0 1 ]
> Con4 B-Amod+ve [ -1 0 0 0 0 1 ]
> Con5 A-Bmod-ve [ 1 0 0 0 0 -1 ]
> Con6 B-Amod-ve [ -1 0 0 0 0 -1 ]
>
> Firstly, is it correct to simply include Con5 and Con6 to obtain
> negative
> correlations, or should I alter the design matrix in some way?
The design matrix is fine but contrasts 3-6 aren't right. A sensible
contrast would be
[0 0 0 0 0 1]
which tests for positive correlation with the behavioural difference.
It doesn't really make sense to include EV1 and EV6 in the same
contrast - the results will be hard to interpret, as you found.
Cheers.
>
>
> Secondly, assuming the above matrix and contrasts are correct, is it
> possible that a small n (e.g. n = 5) could be responsible for FSL
> randomise
> outputting contradictory results that survive cluster thresholding and
> additional Bonferroni correction? For example, I ran randomise using
> the
> above design and contrasts with cluster threshold of 3 and 10000
> permutations. A cluster resulting from the Con4 test overlapped
> (almost
> completely, though it was about 40% the size of) a cluster resulting
> from
> the Con3 test. I checked all_FA_skeletonised voxel values within each
> respective cluster (using fslmeants with -showall option and
> standard stats
> software) and conclude that the Con4 result is incorrect. The other
> results
> seem reasonable, however I should probably abandon the test procedure
> altogether given that randomise has output what seems to me to be at
> least
> one nonsense result.
>
> A similar situation occurred recently with a test group of slightly
> larger
> n, where a cluster from Con6 (negative correlation) overlapped with a
> cluster from Con4 (positive correlation). Again I suspect the small
> n is the
> source of my problem, but I'd be grateful for any further insight or
> suggestions regarding how randomise might be coming up with these
> contradictory results.
>
> Thanks,
>
> Emma
>
---------------------------------------------------------------------------
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director, Oxford University FMRIB Centre
FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
---------------------------------------------------------------------------
