Algorithms underlying flexible phototaxis in larval zebrafish

To thrive, organisms must maintain physiological and environmental variables in optimal ranges. However, in a dynamic world, the optimal range of a variable might fluctuate depending on the organisms state or environmental conditions. Given these fluctuations, how do biological control systems maintain optimal control of physiological and environmental variables? We explored this question by studying the phototactic behavior of larval zebrafish. We demonstrate, with behavioral experiments and computational modeling, that larval zebrafish use phototaxis to maintain environmental luminance at a set point that depends on luminance history. We further show that fish compute this set point using information from both eyes, and that the set point fluctuates on a timescale of seconds when environmental luminance changes. These results expand on previous studies, where phototaxis was found to be primarily positive, and suggest that larval zebrafish, rather than consistently turning towards the brighter areas, exert homeostatic control over the luminance of their surroundings. Furthermore, we show that fluctuations in the surrounding luminance feed back on the system to drive allostatic changes to the luminance set point. Our work has uncovered a novel principle underlying phototaxis in larval zebrafish and characterized a behavioral algorithm by which larval zebrafish exert control over a sensory variable.