Paired T-Test in FSL Using Randomise

I’m attempting to perform a paired T-test in FSL comparing mean differences from scan time 1 to scan time 2 with a within-subjects design. I’m getting an error I can’t resolve.

This is my command:
randomise -i /path/to/4D/data -o /path/to/output/directory -d /path/to/design.mat/file -t /path/to/design.con/file -m /path/to/brain/mask -n 500 -D -T

I receive the following error: ERROR: Program failedMask dimensions do not match input data dimensions!

When I run fslinfo for the input image, it shows the following:
data_type FLOAT32
dim1 91
dim2 109
dim3 91
dim4 38
datatype 16
pixdim1 2.000000
pixdim2 2.000000
pixdim3 2.000000
pixdim4 2.000000
cal_max 0.000000
cal_min 0.000000
file_type NIFTI-1+

When I run fslinfo for the mask image, it shows the following:
data_type UINT8
dim1 364
dim2 436
dim3 364
dim4 1
datatype 2
pixdim1 0.500000
pixdim2 0.500000
pixdim3 0.500000
pixdim4 1.000000
cal_max 250.000000
cal_min 0.000000
file_type NIFTI-1+

My brain mask is the MNI brain, my input image is a 4D volume comprised of the difference maps between condition 1 and condition 2. I’m not sure how to get the input dimensions to match for the input file and the brain mask, especially given that the input file is a 4D volume generated from completely different data than the MNI brain used in the brain mask. Any help?