MRI Acquisition¶

This page documents the MRI acquisition parameters for all LAION-fMRI scans. For the experimental design and task descriptions, see Experimental Design.

Scanner¶

All MRI data were acquired at the Max Planck Institute for Human Cognitive and Brain Sciences in Leipzig, Germany, on a 7T MAGNETOM Terra.X scanner (Siemens Healthineers, syngo MR XA60) with a 7T 8Tx/32Rx head coil (Nova Medical). Participants were scanned in custom-milled styrofoam headcases to reduce head motion, and each subject’s sessions were scheduled within a fixed 2-hour time-of-day window throughout the ~1.5 year acquisition period.

Manufacturer	Siemens Healthineers
Model	SIEMENS MAGNETOM 7.0T W60 Numaris/X VA60A-0CT2
Software	syngo MR XA60
Field strength	7T
Head coil	7T 8Tx/32Rx Head Coil (Nova Medical, Inc.)
Site	Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany

Functional MRI¶

We chose a multi-echo EPI because it allows T2*-weighted optimal echo combination (improving tSNR and BOLD sensitivity) and supports multi-echo ICA denoising via tedana, which uses the TE-dependence of BOLD signal to separate true neural signal from motion and physiological artifacts. For the GLMsingle beta pipeline we work with the optimally-combined (not tedana-denoised) echo to avoid interference with GLMsingle’s own GLMdenoise. The three echo times frame the expected gray matter T2* at 7T (~25-30 ms): TE1 at 11 ms preserves signal in dropout regions, while TE2 and TE3 maximise BOLD contrast in gray matter.

Sequence	gradient-echo multi-band EPI (CMRR sequence), 3 echoes, 2D
TR	1900 ms
TE	TE1 = 11.0 ms, TE2 = 28.82 ms, TE3 = 46.62 ms
Flip angle	60°
Voxel size	1.8 mm isotropic
Matrix size	90 × 110 (base resolution 90, FOV phase 122.2 %)
Number of slices	72
Slice orientation	transverse, tilted -10.4° towards coronal and +1.1° towards sagittal (oblique transverse, roughly aligned to AC-PC plane)
Phase encoding direction	anterior to posterior (A >> P)
Multiband factor	3
GRAPPA	PE acceleration factor 3, FLEET reference scan mode
Partial Fourier	6/8
Fat suppression	yes (Fat Saturation, FA 110°)
Bandwidth	2058 Hz/pixel
Echo spacing	0.59 ms
EPI factor	110
Volumes per run	160 (sessions 1-5) or 163 (sessions 6-34)
Acquisition time per run	5:26 min (sessions 1-5) or 5:32 min (sessions 6-34)
Dummy volumes discarded by scanner	0

Note

The scanner did not discard any volumes before saving, but a 12 second pre-task baseline (~7 volumes at TR = 1.9 s) is included at the start of each run during which no stimulus was shown, to allow signal stabilisation.

Structural MRI¶

High-resolution T2* mapping (MEGRE) was recorded at 7T to improve across-session alignment, but so far it has not been used for later preprocessing. A separate T1w (MP2RAGE) at 3T Siemens Prisma was used for structural alignment of the functional sessions and is described under Retinotopy.

MEGRE (T2*)¶

Sequence	3D multi-echo gradient-echo (MEGRE), monopolar readout, 5 echoes (internal name “fl”)
TR	27 ms
TE	TE1 = 4.0, TE2 = 8.51, TE3 = 13.02, TE4 = 17.53, TE5 = 22.04 ms
TI	not applicable (no inversion pulse)
Flip angle	12°
Voxel size	0.9 mm isotropic
Matrix size	180 × 220 (base resolution 180, FOV phase 122.2 %)
Slices per slab	144
GRAPPA	PE 2 × 3D 2 (24 reference lines each)
Phase partial Fourier	7/8
Slice partial Fourier	7/8
Bandwidth	300 Hz/pixel (same for all 5 echoes)
Distortion correction	2D
Acquisition time	2:55 min (175 s)
Inline T2* map	computed by scanner

Fieldmaps¶

Dual-echo GRE fieldmaps were acquired repeatedly during each session to correct for spatial distortions in the EPI data and to track B0 field drift over time, which can be substantial at 7T.

TR	672 ms
TE	TE1 = 5.0 ms, TE2 = 6.02 ms (ΔTE = 1.02 ms)
Flip angle	50°
Voxel size	1.8 mm isotropic
Number of slices	72
FOV read	160 mm
FOV phase	122.2 %
Base resolution	90
Bandwidth	291 Hz/pixel
Phase partial Fourier	7/8
Distortion correction	2D
Acquisition time per fieldmap	2:10 min (130 s)
Number acquired per session	up to 4 (interleaved through the session to track B0 drift)

Retinotopy & Localizers¶

Todo

Document the acquisition parameters for retinotopy and localizer scans. If they used the same functional sequence as above, just state that. If any parameters differed, note what changed and why.

Diffusion MRI¶

Todo

If DWI data are included: narrative + parameter table (b-values, directions, resolution). If not included, remove this section.