>Date: Wed, 13 Sep 2000 14:05:21 +0100
>To: <[log in to unmask]>
>From: Narender Ramnani <[log in to unmask]>
>Subject: Re: disjunction
>
>Dear Reza,
>
>Presumably you are studying the consolidation of motor learning, very much
>along the lines of your previous studies.
>
>As I understand it, on day 1 you scan subjects on learning (L1) and
>control (C1) tasks. 4 weeks later, you scan again on the same tasks (L2,
>C2), where performance during L1 is enhanced compared with L2 (presumably
>as a result of the intervening consolidation period).
>
>>A reasonable question to ask is what may be the difference between the
>>learning scans across the days, while keeping in mind that there may be
>>non-specific changes that take place from session to session. These
>>non-specific changes might be observed in the activation changes that take
>>place in the control task from session to session.
>
>Your experimental design enables you to look for learning related changes
>that occur over and above non-specific changes. Non-specific factors would
>be common to both L and C. Comparisons of L with C will reveal areas in
>which activity is learning-specific (assuming no motor learning in C). In
>order to find areas in which this difference evolves over time, you should
>look for a 'time-by-condition interaction'. That is, a difference between
>(L2-C2) and (L1-C1).
>
>
>>So, for example, one
>>might look at the difference between the learning scans in day 1 and 2 (call
>>them L1 and L2), and ask if there are voxels there that do not show up in
>>the difference between the control scans between the same two days. This
>>would be: Voxels with significant change in L1-L2 that are not a part of
>>C1-C2. This sounds to me like something that I might be able to do with
>>masking, but I'm not sure. Any thoughts would be appreciated.
>
>The four conditions would simply be modeled as four covariates, so it
>would be more appropriate just to use linear contrasts over the four
>conditions to find the effect. You need not use masking.
>
> If your four covariates were L1 C1 L2 C2, your contrasts would be:
>
>1. -1 1 1 -1 (learning-specific increases in activity)
>2. 1 -1 -1 1 (learning-specific decreases in activity)
>
>Best wishes,
>
>Narender.
********************************************************************
Dr Narender Ramnani
Sensorimotor Control Group
Department of Physiology
University of Oxford
Parks Road
Oxford OX1 3TP
Oxford University Centre for
Functional Magnetic Resonance Imaging of the Brain,
John Radcliffe Hospital,
Headington,
Oxford OX3 9DU
Tel. 01865 222704 (Direct)
01865 222729 (Admin)
mob. 0771 2632785
Fax. 01865 222717
email [log in to unmask]
*******************************************************************
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|