We make a computational simulation of the maturing superior colliculus connected to a simulated facial tissue that replicates some attributes of the bio-mechanical properties of the fetus’ face. We model how the incoming tactile information is used to direct visual attention toward faces. We suggest that the unisensory superficial visual layer (eye-centered) in SC and the deep somatopic layer (face-centered) in SC are combined into an intermediate layer for visuo-tactile integration and that multimodal alignment in this third layer allows newborns to detect faces and to mimic them (see Fig. 1).

Neural populations are defined with integrate-and-fire neurons that capture the spatio-temporal dynamics from the two sensory modalities. The detection of structured patterns is an important attribute for preserving the topology of each modality in each map. The neural populations works similarly to a Kohonen learning systems except that we model the maturing period of SC. We add an activity-dependent mechanism based on novelty detection in order to construct the topology of the neural map by preserving at the same time the existing neurons’ topology and by adding new neurons that refine it.