Social Knowledge Enhances Collective Safety: Computational Models and Simulations

Under common risks (such as mass shootings), people tend to be selfish and not cooperative, which increases the deaths and injuries of civilians. If some are encouraged to fight bravely against the shooters, the deaths and injuries can be reduced substantially. This forms social knowledge, which helps to enhance social safety. However, we cannot use field experiments to calculate social knowledge and evaluate effects. Alternatively, we use an agent-based model to simulate real cases and use counterfactual outcomes to evaluate the beneficial effect of social knowledge. We obtain the optimal solution (with the best matches), based on which we calculate and obtain the social knowledge matrix. Then, we add this matrix into simulations to infer social outcomes. It suggests that social safety has been enhanced substantially: the living chance of shooters has been reduced sharply; the civilian survival rate has been enhanced substantially; and the living chance of heroes themselves has been increased as well. Thus, social safety can be improved, and the effects can be evaluated by computational methods.