Mapping the Development of Spatial Computations from Childhood to Adolescence
Poster No:
2562
Submission Type:
Abstract Submission
Authors:
Jewelia Yao1, Justin Choo1, Dawn Finzi1, Kalanit Grill-Spector1
Institutions:
1Stanford University, Stanford, CA
First Author:
Co-Author(s):
Introduction:
Reading and face recognition, skills essential for children's academic performance and socioemotional development, depend on spatial computations of receptive fields (RFs) – parts of visual space from which information is processed – throughout the ventral visual stream. These behaviors require the use of central vision, relying on RFs of groups of neurons, or population receptive fields (pRFs), which are centrally located (Levy et al., 2001; Hasson et al., 2002; Malach et al., 2002). Despite their significance, few studies have investigated pRFs – especially beyond early visual areas – in children, let alone adolescents. Given the ongoing development of reading and face recognition throughout adolescence, we employ a novel pRF mapping fMRI experiment (Toonoptopy, Finzi et al., 2021) that better drives high-level visual regions, to examine how spatial computations in the ventral visual stream develop from childhood through adolescence into adulthood.
Methods:
19 children (ages 8-17) and 28 adults (ages 22-32) participated in four runs of Toonotopy featuring sweeping bars (5.7º width, Field-of-View: 40º x 40º) filled with cartoon images presented at 8Hz. Each voxel's pRF was modeled using a 2D Gaussian with a compressive nonlinearity (Kay et al., 2013). Thresholded (>20% variance explained) eccentricity, phase, and size maps were generated, and retinotopic ROIs were hand-drawn based on previous definitions for every individual (Wang et al., 2015). pRF parameters (location, size) were estimated for each voxel. In each visual area, pRF parameters, upper visual field bias, the proportion of pRFs in the upper vs. lower visual field; and eccentricity bias, proportion of pRFs in the central 0-5º vs outer 5-20º were compared across development.
Results:
Our approach revealed three main findings. First, 420/444 retinotopic ROIs (V1 – PHC) were identifiable across children and adults for the first time using Toonotopy (Fig 1). Second, linear mixed-effect models revealed little to no development in pRF properties in early visual areas V1 - V3 (Fs < 4, p > .04). Finally, while early visual areas exhibited minimal development, upstream regions continued developing past childhood. Notably, hV4 underwent the most significant change with pRFs becoming more foveal (Fig 2A) across in the right hemisphere (F(1,34) = 13.3, p < 0.0001) and upper visual field bias (Fig 2B) increasing in the right hemisphere (F(1,34) = 4.96, p = 0.03) but decreasing in the left (F(1,35) = 4.37, p = 0.04). Interestingly, in VO pRF size decreased with age in both hemispheres (Fig 2C; LH: F(1, 35) = 6.42, p = 0.02; RH: F(1,33) = 4.76, p = 0.04). Visual field coverage maps also complement these developments, as pRF coverage of the hV4 increases in density in the fovea but decreases in the periphery, and VO coverage spans more of the visual field, from childhood into adulthood.
Conclusions:
Taken together, these findings unveil surprisingly prolonged development through childhood and adolescence of basic spatial computations by population receptive fields in the human ventral visual stream. Importantly, using a novel experimental approach we show differential development across regions and hemispheres whereby intermediate (hV4 and VO) but not early (V1-V3) areas show prolonged development. The fact that basic spatial computations in the visual system continues to develop throughout childhood and adolescence has important ramifications for how the acquisition of centrally-biased visual behaviors crucial for academic and socioemotional growth may affect basic neural computations.
Language:
Reading and Writing
Lifespan Development:
Early life, Adolescence, Aging 2
Modeling and Analysis Methods:
Activation (eg. BOLD task-fMRI)
Segmentation and Parcellation
Perception, Attention and Motor Behavior:
Perception: Visual 1
Keywords:
ADULTS
Cortex
Development
FUNCTIONAL MRI
Language
PEDIATRIC
Perception
Vision
Other - Receptive Fields
1|2Indicates the priority used for review
Provide references using author date format
scaffolded by structural connections,' Nature Communications, 12(1), Article 1. https://doi.org/10.1038/s41467-021-22524-2
Hasson, U. (2002), 'Eccentricity bias as an organizing principle for human high-order object areas', Neuron, 34(3), 479-90. doi: 10.1016/s0896-6273(02)00662-1
Kay, K. N. (2013), 'Compressive spatial summation in human visual cortex. Journal of Neurophysiology', 110(2), 481–494. https://doi.org/10.1152/jn.00105.2013
Levy, I. (2001), 'Center–periphery organization of human object areas', Nature Neuroscience, 4(5), 533–539. https://doi.org/10.1038/87490
Malach, R. (2002), 'The topography of high-order human object areas', Trends in Cognitive Sciences, 6(4), 176–184. https://doi.org/10.1016/S1364-6613(02)01870-3
Wang, L. (2015). 'Probabilistic Maps of Visual Topography in Human Cortex', Cerebral Cortex, 25(10), 3911–3931. https://doi.org/10.1093/cercor/bhu277