Opinion leaders have disproportionate influence in social networks. Watch opinions spread through a network where high-degree nodes seed initial beliefs, competing with random initial opinions.
60
0.50
0.30
—
Opinion A (%)
—
Opinion B (%)
0
Steps
—
Leaders
Model: Each node holds opinion A (blue) or B (red). At each step, each node adopts a neighbor's opinion with probability proportional to neighbor's degree (influence) × (1−stubbornness). Opinion leaders = top-k degree nodes; when seeded, they hold fixed opinions. Scale-free networks (Barabási-Albert) amplify leader effects vs. random networks.