Abstract
How does the functional organization of the ventral visual cortex develop? Most previous studies have concluded that the ventral pathway develops slowly and may take over a decade to fully mature (Golarai et al., 2007; Grill-Spector et al., 2007; Scherf et al., 2007). However, these studies primarily focus on the size and response properties of category-selective regions (e.g., fusiform face area (FFA)). While important, these studies do not shed light on larger-scale representational structures within the visual system. To address this issue, we used representational similarity analysis (Kriegeskorte & Kievit, 2013) to compare the representational structures across the ventral pathways of children and adults. We used fMRI to scan children (ages 5-7, N=39) and adults (N=39) while they viewed videos of faces, bodies, scenes, objects, and scrambled objects. All analyses were carried out in two regions of interest (ROIs): early and ventral visual cortex. Within these ROIs, we correlated the neural activation patterns for each category pairing (e.g., faces and scenes, etc.). This resulted in similarity matrices for each ROI that we then compared across ages by correlating the similarity matrices for each ROI between children and adults. Surprisingly, we found that the similarity matrices between children and adults in ventral visual cortex were nearly identical (r=0.97, p< 0.001), while the similarity matrices in early visual cortex were also very similar between the two groups (r=0.84, p< 0.001). However, the correlations between early and ventral visual cortex both within and between age groups were not significant. These findings, along with prior results, together suggest that even though category-selective may take several years to reach maturation, the large-scale structures across the ventral pathway have already developed by 5-7 years old. These results suggest that large-scale structures develop early and serve as a foundation upon which category-selective regions are built.
Meeting abstract presented at VSS 2016