The recon-all output is in the subject's native space.
So, registration of the volume (MNI) and surfaces (fsaverage) happens as part of the recon-all process, but they're not resampled automatically i.e. kept in the subject's native space. You can use mri_vol2vol command to do convert between spaces.
Recon-all also computes the original volume -> MNI affine transformation, unless the "-no-talairach" argument is passed to recon-all. It's stored in the file: mri/transforms/talairach.xfm. This can be used with mri_vol2vol to go from the native space to the MNI space with either the "--xfm" or "--talxfm" argument:

mri_vol2vol \

--mov <input vol name> \

--targ $FREESURFER_HOME/average/mni305.cor.mgz \

--o <output vol name> \

--talxfm <path to talairach.xfm file>

From: freesurfer-bounces@nmr.mgh.harvard.edu <freesurfer-bounces@nmr.mgh.harvard.edu> on behalf of Andrew M <andrewjmichalak@gmail.com>
Sent: Monday, October 24, 2022 4:32 PM
To: freesurfer@nmr.mgh.harvard.edu <freesurfer@nmr.mgh.harvard.edu>
Subject: [Freesurfer] Recon-all output in MNI305, patient space, or other?

External Email - Use Caution

Greetings,

I am running recon-all as per the iElvis pipeline with:

recon-all -s <id> -i $SUBJECTS_DIR/.../T1.nii -all -localGI

And followed by ct2mri

My questions are:

1. What is the actual output "space" of the resulting T1.nii and postInPre.nii ? Does recon-all do keep everything in patient space, or does it register them to fsaverage space automatically? If so, how would I convert these MRIs and CTs to MNI305 space?

Thank you very much.

Andrew