Behavioral algorithms of ontogenetic switching in larval and juvenile zebrafish phototaxis

Animals undergo major behavioral adjustments during ontogeny, but how the underlying cognitive algorithms change during this process remains elusive. Here, we describe that zebrafish shift from light-seeking to dark-seeking as they grow from larval to juvenile stage. Combining complementary phototaxis assays in virtual reality and modeling, we dissect the computational basis of this transition. We identify three parallel pathways, one analyzing ambient whole-field luminance levels, one spatially comparing light levels across the eyes, and one computing luminance change in each eye. Larvae mostly use the latter two computations, whereas juveniles largely employ the first one. Using a library of agent-based models, we predict behavior in more complex environments. Model-based extraction of latent cognitive variables suggests potential neural correlates of this behavioral inversion. We suggest that zebrafish phototaxis is regulated via parallel processing streams, which could be a universal mechanism for adjusting behavior depending on developmental stage, context, or internal state.