Fig. 2: Critical mass dynamics.
From: Group interactions modulate critical mass dynamics in social convention
a Illustrative example of a simulation of the Naming Game (NG) with unanimity rule on an empirical social structure (Thiers13), where a minority Ac of one single committed individual supporting A--consisting of 0.3% of the population of 327 individuals--overturns the stable social norms and reaches global consensus (under imperfect communication, with social influence parameter β = 0.336). b Temporal evolution of the fraction nx(t) of nodes supporting name x. Different solid lines correspond to different names, x = {A + Ac, B, (A, B)}. Dashed lines are reported as a benchmark, representing the case with perfect communication (β = 1). c Temporal evolution of the normalised size of the largest connected component (LCC) of nodes supporting name A (red curve) and nodes that have A (but not necessarily A only) in their vocabulary (green curve). Panels d, e, f, g show the temporal evolution of the dynamics with committed minorities (p = 3%) on empirical higher-order structures. The social structures are constructed from empirical data sets collected in six different context: a workplace (InVS15)84, a primary school (LyonSchool)85, a conference (SFHH)86, a high school (Thiers13)58, email communications (Email-EU)87 and a political congress (Congress-bills)88. The temporal evolution of the densities of nodes holding name A and holding both A and B are reported in panels (d, e) and (f, g), respectively, for two different values of the parameter β quantifying the efficacy of reaching an agreement within a group, namely β = 0.28 (d, f) and β = 0.41 (e, g). The results over different runs of stochastic simulations are reported as median values (solid lines) and values contained within the 25th and 75th percentiles (shaded areas).
