| >	1) Also use a symmetric weighting image for the spatial normalisation.
| >	   This can be specified by modifying the spm_defaults.m  file at
| >	   around line 191 so that it uses a weighting image that has also
| >	   been left-right flipped and averaged.
| 
| I understand this can be done by setting sptl_MskObj = 1
| in the spm_defaults.m file. But I am not sure which image
| should be chosen as "object masking image". An average
| image between L-R flipped and unflipped images of the
| object? Should this image scaled so that the contained
| values will be between 0 and 1?

Not quite.  The symmetric weighting image is the one that is currently specified
as default by:
	sptl_MskBrn = fullfile(spm('Dir'),'apriori','brainmask.img');

The apriori/brainmask.img that is released with SPM99 is not symmetric.

| >	2) Remember that rotations, zooms and translations are also included
| >	   in the deformation fields - not just shape.  In order to discount
| >	   the effects of subject positioning in the scanner, you will need
| >	   to somehow normalise the deformation fields for subject position.
| >	   Also think about whether you want to normalise out head size from
| >	   the deformations.
| 
| These factors can be discounted by applying two-step
| normalization, namely, performing affine transformation
| to all images first, and then performing non-linear
| normalization on these affine-transformed images.
| Correct?

This is probably not the best way.  Discretely tucked away in the SPM99
distribution, you will find a function called spm_dbm.m (which is not
explicitly invoked by any of the buttons within SPM99).  Between lines
258 and 390 is a piece of code for factoring out translations, rotations
and an isotropic zoom from a deformation field.  The deformation field
that remains is then parameterised in terms of its DCT coefficients.
There should be something in this piece of code that is helpful to you.

| Great. Yet, as far as I can see, the basic stat menu has
| only 1-way option for MANOVA. Are there any tricks to
| implement 2-way MANOVA with this routine?

Not sure.

| 
| >	   Note that a Wilk's Lambda field transformed to a Chi^2 field is
| >	   not exactly the same as a pure Chi^2 field.  The corrections for
| >	   numbers of resels is therefore not exact, but I'm sure it's still
| >	   very close.
| 
| Just a casual idea. In special cases of 1-way MANOVA,
| a Wilk's Lambda field can be exactly transformed to a
| F field (Johnson & Wichern, Applied multivariate
| statistical analysis, 4th ed., 1998, p323). I guess
| Worsley's theory can be exactly applied in those cases.
| (I am not a statistician. Please someone correct me if
| wrong.)

I was wrong about this one.  I have just read the code, and it appears to be
transformed to an F statistic rather than a Chi^2.  The documentation says:
	% MV:
	% If the response variable is multivariate (i.e. size(VY,2) > 1) then
	% spm_spm proceeds with a voxel by voxel ManCova to produce a SPM{F}
	% based on Wilks Lambda for all effects of interest (specified by F_iX0).
	% The ensuing parameter estimates, data (Y.mad) and residual sum of squares
	% pertain to the first canonical variate. This is the linear combination
	% of response variables that maximizes the sum of squares explained
	% by the effects of interest relative to error.  Because there is only
	% one contrast (F_iX0) the SPM{F} is created at this point and details
	% are saved in xCon.

I know that a field of t-statistics that have been transformed to Z-statistics
is not a Z-field, so I was assuming that a Wilk's Lambda field transformed to
a Chi^2 (or indeed an F) does not behave exactly like a Chi^2 field.  I am not a
statistician either, so someone please correct me if I am wrong.

Best regards,
-John



%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%