Effect of Suppression Width in Spontaneous Detection of Numerosity

Abstract

Number sense is the spontaneous ability of human adults, infants, newborns, and other animals to process approximate numbers. However, how number sense can emerge in the absence of learning and fine-tuning has not been elucidated. On the basis of the treatment of visual pathways as an information processor, we developed two computational models for numerosity detection with single and dual suppression values. The information processor transforms the visual information of numerosities into input information of neurons in the parietal and prefrontal cortex via suppression and visuotopic mapping without the need for the construction of neural networks, training, and fine-tuning. In the single suppression model, the suppression width is a more fundamental determinant than numerical quantity in numerosity detection. Our dual suppression model clarifies the effect of suppression width values in two visual regions on spontaneous numerosity detection. It also predicts innate and acquired training behavior in humans and animals. Results suggest that the range of the lateral inhibition of neurons at successive stages in visual pathways is the neural basis of human number sense that ultimately gives rise to number theory and mathematics.