Competing for influence in networks through strategic targeting

We experimentally investigate targeting decisions in a setting where a human player competes for influence in a network against a computerized opponent with opposing views, whose targeting choice is revealed before the player acts. By varying network structure, opponent influence, and nodes opinion heterogeneity, we find that players typically adopt best-response strategies based on relative influence. However, they sometimes deviate – for example, by erroneously targeting central nodes or by avoiding the opponent’s target. Targeting is also affected by affinity and opposition biases, the strength of which depends on the initial opinion distribution. Targeting the center, avoiding the competitor’s target, or selecting nodes based on their initial opinions when these are not best responses generates significant efficiency losses.