Decoding the directionality of information flow in brain circuits using ultrafast fMRI
Poster No:
1458
Submission Type:
Abstract Submission
Authors:
Joana Carvalho1, Francisca Fernandes2, Mafalda Valente2, Koen Haak3, Noam Shemesh2
Institutions:
1Faculty of Psychology and Educational Sciences, University of Coimbra, Coimbra, Coimbra, 2Champalimaud Foundation, Lisboa, Lisboa, 3Department of Cognitive Science and Artificial Intelligence, Tilburg University, Tilburg, Tilburg
First Author:
Joana Carvalho
Faculty of Psychology and Educational Sciences, University of Coimbra
Coimbra, Coimbra
Faculty of Psychology and Educational Sciences, University of Coimbra
Coimbra, Coimbra
Co-Author(s):
Koen Haak
Department of Cognitive Science and Artificial Intelligence, Tilburg University
Tilburg, Tilburg
Department of Cognitive Science and Artificial Intelligence, Tilburg University
Tilburg, Tilburg
Introduction:
Sensory processing relies on hierarchical, bidirectional communication between early sensory and higher cortical areas, essential for understanding brain dynamics, learning, and neuroplasticity. No current non-invasive technique can distinguish feedforward (FF) from feedback (FB) signals across entire networks, including deep brain regions. Here, we developed a Layer-based Connective Field (lCF) model using ultra-high spatiotemporal resolution fMRI (task-based and resting-state-RS) to differentiate layer-specific FF and FB signaling in healthy and brain-injured animals.
Methods:
Experiments were preapproved by the authorities. Twenty Long Evans rats underwent two sessions: Session 1 for retinotopic mapping and Session 2 for ultrafast acquisitions. Three ultrafast experimental sets explored FF and FB connectivity in the visual system: (1) standard spatial resolution resting-state (RS) scans (228 µm x 228 µm, TR=350ms/50ms), (2) high spatial resolution scans (RS and stimulation, 120 µm x 120 µm, TR=50ms), and (3) RS scans in rats with bilateral V1 lesions. Additionally, 6 rats were scanned in somatomotor and motor pathways. Images were NORDIC denoised, slice-timing and motion corrected, coregistered, normalized to the SIGMA atlas [1]. Population receptive field mapping and layer Connective field model (lCF) builds on the work of [2,3]
Results:
We targeted FF and FB pathways in the rodent visual system, focusing on the geniculate pathway (LGN → V1 : layer IV) and the extrageniculate pathway (LP → HVAs: layers IV and V; Figure 1A). The CF center indicates the location of highest correlation, while its size reflects information integration, capturing topographically specific BOLD responses [4,5]. Figure 1C-D shows retinotopic CF projections with phase reversals aligning to visual area borders, suggesting visuotopic organization persists without visual input. LGN projections to V1 exhibit large CF sizes in layer IV with a ∩-shaped profile tapering in superficial and deeper layers (Figures 1E-G), highlighting precise targeting. This ∩-shaped pattern is absent in LGN → V2 projections, which show a linear CF size increase with depth.
LPN projections, primarily targeting HVAs, display a ∩-shaped profile in V2, peaking in middle layers (Figure 2H), while LP → V1 projections show a linear increase similar to LGN → V2. This linear trend may result from noise in deeper cortical layers. ANOVA analysis revealed significant differences in CF profiles for LP → V2 and LGN → V1 compared to LP → V1 (F(1,10)=14.68, p=0.003) and LGN → V2 (F(1,10)=8.57, p=0.015), but not between LP → V2 and LGN → V1 or LP → V1 and LGN → V2 (Figure 1I).
Using single slice acquisitions with a ultrahigh spatiotemporal resolution during spontaneous and continuous binocular visual stimulation (flickering light), we computed layer-to-layer CFs between V1 and V2 layers (Figure 2A). Layer-to-layer CF sizes revealed two distinct connectivity profiles across cortical layers that align with layer-specific FF and FB patterns: V1 projects information to the middle layers of V2 (∩-shaped profile), while deep layers of V2 predominantly relay FB information to the superficial and deep layers of V1 (U shaped profile, Figure 2).
Importantly, our findings extended beyond the visual system, revealing FF- and FB-specific CF patterns in the somatomotor and motor pathways.
Additionally, we show a first application in cortically blind animals, revealing that higher visual areas receive direct FF input from the LGN, bypassing V1 and following a ∩-shaped CF size pattern.
LPN projections, primarily targeting HVAs, display a ∩-shaped profile in V2, peaking in middle layers (Figure 2H), while LP → V1 projections show a linear increase similar to LGN → V2. This linear trend may result from noise in deeper cortical layers. ANOVA analysis revealed significant differences in CF profiles for LP → V2 and LGN → V1 compared to LP → V1 (F(1,10)=14.68, p=0.003) and LGN → V2 (F(1,10)=8.57, p=0.015), but not between LP → V2 and LGN → V1 or LP → V1 and LGN → V2 (Figure 1I).
Using single slice acquisitions with a ultrahigh spatiotemporal resolution during spontaneous and continuous binocular visual stimulation (flickering light), we computed layer-to-layer CFs between V1 and V2 layers (Figure 2A). Layer-to-layer CF sizes revealed two distinct connectivity profiles across cortical layers that align with layer-specific FF and FB patterns: V1 projects information to the middle layers of V2 (∩-shaped profile), while deep layers of V2 predominantly relay FB information to the superficial and deep layers of V1 (U shaped profile, Figure 2).
Importantly, our findings extended beyond the visual system, revealing FF- and FB-specific CF patterns in the somatomotor and motor pathways.
Additionally, we show a first application in cortically blind animals, revealing that higher visual areas receive direct FF input from the LGN, bypassing V1 and following a ∩-shaped CF size pattern.
Conclusions:
In this study, we disentangled FF and FB patterns using fMRI for the first time, leveraging high spatial and temporal resolution acquisitions and bidirectional computational models of cortical information flow. This non-invasive tracking of FF and FB signals is essential for understanding brain plasticity and the interactions between brain regions during sensory processing, decision-making, and cognitive tasks.
Modeling and Analysis Methods:
Connectivity (eg. functional, effective, structural)
fMRI Connectivity and Network Modeling 1
Task-Independent and Resting-State Analysis 2
Novel Imaging Acquisition Methods:
BOLD fMRI
Perception, Attention and Motor Behavior:
Perception: Visual
Keywords:
ANIMAL STUDIES
Computational Neuroscience
Cortical Layers
FUNCTIONAL MRI
HIGH FIELD MR
Modeling
Plasticity
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
For human MRI, what field strength scanner do you use?
Which processing packages did you use for your study?
Provide references using APA citation style.
2. Haak, K. V. et al. Connective field modeling. Neuroimage 66, 376–384 (2013).
3. Dumoulin, S. O. & Wandell, B. A. Population receptive field estimates in human visual cortex. Neuroimage 39, 647–660 (2008).
4. Gravel, N. et al. Cortical connective field estimates from resting state fMRI activity. Front. Neurosci. 8, (2014).
5. Knapen, T. Topographic connectivity reveals task-dependent retinotopic processing throughout the human brain. Proc Natl Acad Sci U S A 118, (2021).