Hi,
No, I'm saying that for _contrasts_ 3-6 you should remove _EV_ 1 - if
you want to test for correlation with the additional covariate, you
shouldn't include the main group difference (EV1) in that contrast.
If you want a paired design (which is equivalent to pre-subtracting
each pair) then I don't see how you can easily separate the
behavioural correlation into A vs B - you can just look at
(effectively) the modulation of the difference by the covariate -
though maybe Tom has an alternative idea.......
Cheers.
On 27 Nov 2008, at 07:56, Emma Bendall wrote:
> 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
> ---------------------------------------------------------------------------
>
---------------------------------------------------------------------------
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
---------------------------------------------------------------------------
|