Are bridging ties really advantageous? An experimental test of their advantage in a competitive social learning context

Despite the widespread acceptance of the claim that bridging ties help to obtain profitable outcomes, its underlying mechanisms remain understudied. Starting from a multi-armed bandit problem, we tested the bridging tie hypothesis experimentally by studying the outcomes of social learning for different network positions (in terms of local clustering and closeness centrality) with and without competition. We found a positive effect of bridging ties, but only within one's direct network (i.e., when local clustering is lower), in competitive contexts, and for choices characterized by higher uncertainty. This stresses the importance of outlining more clearly the scope in which the bridging tie hypothesis applies.