Characterizing individual top-down depth-dependent human microcircuitry in ventro-temporal cortex
Poster No:
2541
Submission Type:
Abstract Submission
Authors:
Luca Vizioli1, Logan Dowdle1, Geoff Ghose1, Steen Moeller1, Kamil Uğurbil1, Essa Yacoub1
Institutions:
1Center for Magnetic Resonance Research, Minneapolis, MN
First Author:
Co-Author(s):
Introduction:
Feedforward and feedback connections between different cortical areas define the sophisticated hierarchical processing of information in neocortex and have been characterized by distinct patterns of laminar connectivity using invasive methods in animals. Recent MR technological advances have enabled acquiring functional images with submillimeter resolution over small volumes, allowing noninvasive access to this mesoscopic scale. However, studying laminar connectivity non-invasively in humans remains challenging as extending these methods to large volumes encompassing multiple areas, critical for establishing the connectivity between regions, has remained elusive as the high level of acceleration required to achieve large coverage at high resolutions is restricted by thermal noise effects. Here we record functional images simultaneously from several regions of varying complexity across the cortical hierarchy to examine the impact of top-down task modulations during a high-level task.
Methods:
We recorded 7T BOLD images in 4 individuals (Gradient echo (GE) EPI; TR: 2.03s; R=3; MB=2; 6/8th Partial Fourier; TE: 29.6ms; 56 slices). In two tasks, subjects viewed 8 blocks (12s on/off duration) of face stimuli. Faces were phase scrambled between 0 to 40% coherence1. Participants completed 6 runs of a face perception and 6 of a stimulus irrelevant fixation task. Before standard preprocessing (i.e., slice-timing; motion and detrending) images were denoised with NORDIC, which suppresses thermal noise without impacting functional spatial precision2. To identify face preferential regions such as the Fusiform face area (FFA) and Occipital face area (OFA), we used a separate face localizer. The retinotopic representation of the stimuli in V1 was instead localized using the first block of each task (discarded from subsequent analyses) in conjunction with probabilistic atlases. The regions of interest (ROIs) were converted into 3 cortical depths using LayNii3. Activation was estimated for each of the 48 single blocks, allowing enough statistical power to carry out within subject statistical analyses. To evaluate top-down effects of task demands, we calculated the ratio of the activation during the face relative to the fixation task for each cortical depth. We then performed a depth dependent functional connectivity analysis, but in this case, it was done independently per task.
Results:
Within each subject we show that, relative to the fixation, the face task elicited significantly larger BOLD responses in the inner compared to the outer layers of V1; and in the outer compared to the inner layers in the FFA (bootstrapped p<0.05, Bonferroni corrected). Moreover, we report a significant (p<.05, Bonferroni corrected) increase in functional connectivity between outer depths of FFA and inner depths of V1, during the face relative to the fixation task.
Conclusions:
Cortical depth dependent fMRI allows non-invasive mesoscopic investigations of behaviors that are unique to humans. However, due to SNR and imaging challenges, prior studies have been mostly confined to primary sensory cortices using low-level stimuli, which do not tap into human specific behaviors. Here, we examine top-down cortical depth dependent modulations concomitantly across areas of different complexity in the visual hierarchy during a high level, socially meaningful task. Our layer dependent findings are consistent with animal reports of feedback being exchanged between deeper and superficial layers4. Furthermore, these results are consistent with a theorized mechanism of conscious perception involving apical dendritic amplification5. Importantly, the dissociation in laminar profiles between the FFA and V1 indicates that the results cannot be accounted for by spurious venous effects. Our work demonstrates the feasibility of carrying out depth dependent fMRI analyses at the single subject level, a crucial step towards characterizing mesoscopic differences in the diseased brain.
Modeling and Analysis Methods:
Activation (eg. BOLD task-fMRI)
fMRI Connectivity and Network Modeling 2
Perception, Attention and Motor Behavior:
Perception: Visual 1
Perception and Attention Other
Keywords:
Cortical Layers
FUNCTIONAL MRI
Perception
Vision
1|2Indicates the priority used for review
Provide references using author date format
2. Vizioli, L. et al. (2021) Lowering the thermal noise barrier in functional brain mapping with magnetic resonance imaging. Nat Commun 12, 5181
3. Huber, L. (Renzo) et al. (2021) LayNii: A software suite for layer-fMRI. NeuroImage 118091 doi:10.1016/j.neuroimage.2021.118091.
4. Rockland, K. S. & Pandya, D. N. (1979) Laminar origins and terminations of cortical connections of the occipital lobe in the rhesus monkey. Brain Research 179, 3–20.
5. Marvan, T., Polák, M., Bachmann, T. & Phillips, W. A. (2021) Apical amplification—a cellular mechanism of conscious perception? Neuroscience of Consciousness.