1. Summary

Scrutiny of the cortical neuronal circuits underlying human visual perception typically involves the summarization of large-scale recordings of brain activity under different perceptual states, with the combination of various measurement modalities and modeling techniques being critical in revealing organizing principles. However, how these approaches relate to the brain’s anatomical structure at a more local scale, such as that of retinotopically organized neuronal circuits in the visual cortex, remains an area of active research. Here, we set out to study such circuits by applying a previously established model of network communicability 1 to resting state (RS), movie watching (MOVIE) and visual field mapping (VFM) data from 154 subjects in the Human Connectome Project (HCP) 7T-fMRI dataset 2. By estimating the effective connectivity (EC) of the noise-diffusion network model that better predicts the spatiotemporal covariance structure of the data, we disentangle the contributions of directed interactions and local dynamics in shaping blood-oxygen dependent (BOLD) activity propagation across visual cortical areas V1, V2 and V3 3. First, addressing the need for validation of our approach, we confirm previous findings using 7T-fMRI and a smaller dataset for evidence of task-dependent changes in directed interactions across early visual cortex 4. We then extend these findings by including a new task (MOVIE) and show that a similar reconfiguration of feedforward and feedback connections between parafoveal and foveal regions occur, with feedback- outweighting feedforward- interactions during cognitively demanding tasks (VFM and MOVIE). Our findings demonstrate the reliability of our approach to summarize the neuronal interactions unfolding across the underlying circuit layout of the visual cortex during different perceptual states. We conclude that the relationship between BOLD statistics and the propagation of neuronal activity summarized by our modeling approach is of fundamental and methodological interest, as it opens the door to study neuronal circuit dynamics during different tasks non-invasively in humans.

1.1. References

[1]

Matthieu Gilson, Gustavo Deco, Karl J. Friston, Patric Hagmann, Dante Mantini, Viviana Betti, Gian Luca Romani, and Maurizio Corbetta. Effective connectivity inferred from fmri transition dynamics during movie viewing points to a balanced reconfiguration of cortical interactions. NeuroImage, 180:534–546, 2018. Brain Connectivity Dynamics. URL: https://www.sciencedirect.com/science/article/pii/S1053811917308108, doi:https://doi.org/10.1016/j.neuroimage.2017.09.061.

[2]

Noah C. Benson, Keith W. Jamison, Michael J. Arcaro, An T. Vu, Matthew F. Glasser, Timothy S. Coalson, David C. Van Essen, Essa Yacoub, Kamil Ugurbil, Jonathan Winawer, and Kendrick Kay. The Human Connectome Project 7 Tesla retinotopy dataset: Description and population receptive field analysis. Journal of Vision, 18(13):23–23, 12 2018. URL: https://doi.org/10.1167/18.13.23, arXiv:https://arvojournals.org/arvo/content\_public/journal/jov/937687/i1534-7362-18-13-23.pdf, doi:10.1167/18.13.23.

[3]

Nicolás Gravel, Remco J Renken, Ben M Harvey, Gustavo Deco, Frans W Cornelissen, and Matthieu Gilson. Propagation of BOLD Activity Reveals Task-dependent Directed Interactions Across Human Visual Cortex. Cerebral Cortex, 30(11):5899–5914, 2020. doi:10.1093/cercor/bhaa165.

[4]

Mark M. Schira, Christopher W. Tyler, Branka Spehar, and Michael Breakspear. Modeling Magnification and Anisotropy in the Primate Foveal Confluence. PLoS Computational Biology, 6(1):e1000651, 2010. doi:10.1371/journal.pcbi.1000651.