Abstract:

The left visual word form area (VWFA) of the brain in adults is more sensitive to orthographic information, whereas the right fusiform face area (FFA) is preferentially involved in the processing of facial information. The cognitive neural processing mechanism of the competitive development of the complementary pattern of hemispheric lateralization in word and face recognition is systematically examined using the neuronal recycling hypothesis and the distributed account of the hemispheric organization, combined with the multidimensional computational model composed of posterior-anterior axis and medial-lateral axis in the fusiform gyrus (FG). The posterior-anterior axis distinguishes regions within a domain and is associated with hierarchical transformation. The medial-lateral axis differentiates regions across domains. Along the posterior-anterior axis in FG, area FG2, which is posterior lateral to FG, receives mainly bottom-up incoming information from the early visual cortex. Additionally, both word and face recognition require fine discrimination between highly similar stimuli, relying on high-acuity visual information from the fovea. Therefore, word and face representation should share processing resources for early perception and compete for neural space in FG2. These areas are sculpted further over development and begin to emerge being labelled the VWFA-1 in the left hemisphere (lateral to the medial-lateral axis) and the FFA-1 in the right hemisphere (medial to the medial-lateral axis). Moreover, as learning to read words and face recognition experience accumulated, the processing of word or face information transformed from VWFA-1 and FFA-1 in FG2 to VWFA-2 (lateral to the medial-lateral axis) and FFA-2 (medial to the medial-lateral axis) in FG4 located in the lateral middle part of FG, respectively. Because learning to read words improves the connectivity between VWFA-2 and the left hemisphere language network, this top-down lateralization modulation strengthens the competitive processing between words and faces, resulting in stronger leftward lateralization of VWFA-2 and accelerating rightward lateralization of FFA-2. At the same time, the face learning experience also improves the right-sided connectivity between FFA-2 and related brain networks, and this top-down modulation results in stronger right-lateralized FFA-2.

Future studies should examine the following five aspects. First, some researchers propose a revised model of the neuronal recycling hypothesis, known as the blocking model, in which they suppose that in literate children, the progressive dedication of an increasing number of neuronal patches to words in the left FG prevents the expansion of the nearby object and face patches, resulting in face recognition being gradually replaced by right FG. Therefore, future studies could use high-resolution (<1mm) functional magnetic resonance imaging (fMRI) to examine whether words directly compete with faces or block the development of faces. Second, while cytoarchitectonic similarities between FG2 and FG4 are found, the differences between them may help explain the fine-scale functional heterogeneity of FG. Based on differences in cell types, receptor densities, laminar distribution patterns, synaptic connections, and myelination between FG2 and FG4, further research could provide insight into the functional role of both fusiform areas subserving face and word recognition. Third, research has shown that preschool children's learning to read Chinese characters requires the right hemisphere to distinguish character form, and use the left hemisphere to access orthography and phonological and semantic information, so their early reading experience may compete with faces for processing resources in the right and left hemispheres. It is necessary to adopt the intertemporal study design of preschool and school-age children and use event-related potentials (ERPs) with high temporal resolution and fMRI with high spatial resolution to systematically investigate the unique mechanism of children's early reading experience affecting the competitive processing of Chinese characters and faces. Fourth, the development of right-lateralized face recognition may be influenced by factors other than learning to read words. The development of right-lateralized face recognition will experience a long time delay, which is influenced not only by brain development and learning to read but also by the gradual accumulation of facial experience. Therefore, future studies should not only focus on how learning to read words competes with face representation for processing resources in left FG, but also examine how face perception experience affects the right-lateralized development of FFA and its processing mechanism in competition with other cortical areas for space resources. Finally, visual numbers and musical notations are also cultural inventions, and learning to read them can cause brain plasticity changes similar to words. The brain structures and functional networks of learning to read words, numbers, and musical notations are both separated and overlapped, competing and coordinating with each other. Importantly, children can learn to read these three visual symbols synchronously or asynchronously. Therefore, exploring the temporal dynamics and brain mechanisms of the interaction between children's learning to read words, numbers, and musical notations not only has theoretical significance for deepening the understanding of brain plasticity of cultural learning, but also has practical significance for guiding the early teaching of language, mathematics, and music.

Key words: visual word form area, fusiform face area, complementary pattern of hemispheric lateralization, mechanism of competitive development, multilevel and bidirectional dynamic processing