A studyforrest extension, retinotopic mapping and localization of higher visual areas

Abstract

The studyforrest (http://studyforrest.org) dataset is likely the largest neuroimaging dataset on natural language and story processing publicly available today. In this article, along with a companion publication, we present an update of this dataset that extends its scope to vision and multi-sensory research. 15 participants of the original cohort volunteered for a series of additional studies: a clinical examination of visual function, a standard retinotopic mapping procedure, and a localization of higher visual areas-such as the fusiform face area. The combination of this update, the previous data releases for the dataset, and the companion publication, which includes neuroimaging and eye tracking data from natural stimulation with a motion picture, form an extremely versatile and comprehensive resource for brain imaging research-with almost six hours of functional neuroimaging data across five different stimulation paradigms for each participant. Furthermore, we describe employed paradigms and present results that document the quality of the data for the purpose of characterising major properties of participants' visual processing stream.

Figure 1. Retinotopic mapping validation.

(a) Ring and wedge stimuli with continuous central letter reading task to encourage fixation. White numbers indicate the respective phase angle encoding. (b) Histogram of polar angles for all voxels in the MNI occipital lobe mask for the left and right hemisphere. Error bars indicate s.d. across all subjects. (c) Inflated occipital cortex surface maps for eccentricity and polar angle for the best, intermediate, and worst participants: participants 1, 10, and 9 respectively. White lines indicate manually delineated visual area boundaries; stars mark the center of the visual field; yellow lines depict the outline of the autogenerated Freesurfer V2 label for comparison. All maps are constrained to the MNI occipital lobe mask.

Figure 2. Experiment design.

(a) In each block, 16 unique images were presented on a medium-gray background (with a superimposed green fixation cross). Each image was shown for 900 ms and images were separated in time by 100 ms. The participant’s task was to press a button (index finger, right hand) when any image was immediately followed by its mirrored equivalent. These events happened randomly either once, twice, or never in each block. In order to alert the participant, the fixation cross turned green 1.5 s prior to the start of a block, remained green throughout a block, and was white during the rest period. The start of each block was synchronized with the MR volume acquisition trigger pulse. Stimulus blocks were separated by 8 s of fixation. (b) Example images for all six stimulus categories.

Figure 3. Average temporal signal-to-noise ratio (tSNR) across all acquisitions, including the fMRI data described in the companion article.

tSNR was computed independently from motion-corrected and linearly detrended BOLD fMRI time series for each scan. The resulting statistics were projected into group space for averaging across scans and participants (n=255). (a) Spatial distribution of average tSNR across the brain. In the vicinity of the ventricles, tSNR is reduced due to a suboptimal fat suppression SPIR pulse frequency. This artifact amplifies the expected U-shape of the spatial SNR profile of a 32 channel head coil. (b) Histograms of average tSNR scores. The dark shaded histogram shows the tSNR distribution of all voxels in an approximate brain mask (MNI template brain mask); the lighter shaded histogram shows the tSNR distribution of 20% of voxels with the largest probability of sampling gray matter, as indicated by FSL’s gray matter prior volume (avg152T1_gray.nii.gz) for the MNI template image.

Figure 4. Spatial overlap of individually located regions of interest (ROI).

Individual ROI masks were projected into MNI space and summed. Absolute value are shown for nodes on the reconstructed surface of the MNI152 brain. The magnitude is affected by both the spatial variability of ROIs across brains and the localization failures for individual ROIs and participants. See Tables 2 –4 for individual results.