Precise visuomotor transformations underlying collective behavior in larval zebrafish

Complex schooling behaviors result from local interactions among individuals. Yet, how sensory signals from neighbors are analyzed in the visuomotor stream of animals is poorly understood. Here, we studied aggregation behavior in larval zebrafish and found that over development larvae transition from overdispersed groups to tight shoals. Using a virtual reality assay, we characterized the algorithms fish use to transform visual inputs from neighbors into movement decisions. We found that young larvae turn away from virtual neighbors by integrating and averaging retina-wide visual occupancy within each eye, and by using a winner-take-all strategy for binocular integration. As fish mature, their responses expand to include attraction to virtual neighbors, which is based on similar algorithms of visual integration. Using model simulations, we show that the observed algorithms accurately predict group structure over development. These findings allow us to make testable predictions regarding the neuronal circuits underlying collective behavior in zebrafish.


nature research | reporting summary
April 2020 Field-specific reporting Please select the one below that is the best fit for your research. If you are not sure, read the appropriate sections before making your selection.

Life sciences Behavioural & social sciences Ecological, evolutionary & environmental sciences
For a reference copy of the document with all sections, see nature.com/documents/nr-reporting-summary-flat.pdf

Life sciences study design
All studies must disclose on these points even when the disclosure is negative. Wild animals

Field-collected samples
For all group swimming experiments we used sample sizes that were large enough to estimate group statistics (e.g. dispersion and alignment) according to previously reported data on collective behavior in zebrafish (see for example Katz et al, 2011, Harpaz et al, 2017 and Methods for statistical procedure and references) and to also allow at least 25 degrees of freedom when parametric statistical models were used to compare between experimental conditions. In the virtual reality assay, we used 40 trials per stimulus as this number proved sufficient to estimate the response of a single fish to the presented stimuli (see also Larsch et al, 2018) and 24-32 fish were used per experiment as our preliminary data showed that these numbers are sufficient to estimate the mean and variance of the responses of the fish to the presented stimuli, and to reliably detect differences between stimuli.
Groups were eliminated from subsequent analysis in the case that one or more of the fish were immobile for more than 25% of the experiment. This criteria was decided based upon preliminary observations, but we have also verified that choosing a more stringent, or a less stringent criteria for elimination did not change the nature of the results. In the VR assay, fish were eliminated and replaced before stimuli presentation begun if they did not show spontaneous swimming for more than 5 minutes.
All experimental results presented here were successfully replicated. We also chose to test two different group sizes (5 and 10 fish in a group) to make sure our results can generalize to groups of different sizes. In the VR assay we show that the the behavioral algorithms we detected and report, also generalized to different stimuli (dots vs ellipses, moving vs. stationary stimuli).
For all group swimming experiments, fish (of the same age group) were randomly assigned to groups of 5 or 10 fish. In the VR assay, individual fish were randomly assigned to an experiment (see manuscript for different experiments and stimuli). The order of stimuli presentation was randomly shuffled for every fish.
Experimenters were not blinded to the experimental conditions. For the group swimming experiments, the experimental conditions that we manipulated were age and light vs dark which are directly controlled by the experimenter. For the VR assay, all fish in an experiment were presented with all stimuli and therefore blinding was unnecessary.
Zebrafish, Danio rerio of the AB strain at ages 7, 14 and 21 dpf. Sex is not fully determined at these ages.
Provide details on animals observed in or captured in the field; report species, sex and age where possible. Describe how animals were caught and transported and what happened to captive animals after the study (if killed, explain why and describe method; if released, say where and when) OR state that the study did not involve wild animals.
For laboratory work with field-collected samples, describe all relevant parameters such as housing, maintenance, temperature, photoperiod and end-of-experiment protocol OR state that the study did not involve samples collected from the field.