On the speed-accuracy tradeoff in collective decision making

We study collective decision making in human groups performing a two alternative choice task. We model the evidence aggregation process across the network using a coupled drift diffusion model (DDM) and consider the free response paradigm in which humans take their time to make the decision. We analyze the coupled DDM under a mean-field type approximation and characterize approximate error rates and expected decision times for each individual in the group as a function of their location in the network. We also provide approximations to the first passage time distributions for each individual. We elucidate criteria to select thresholds for decision making in human groups as well as in engineering applications.