Weak Ties, Wide Reach: A Learning Model Breaks Echo Chambers
A Northwestern University study introduces a learning-augmented network model showing that positive reinforcement strengthens existing connections and creates ‘death spirals’ that trap activity, while negative reinforcement weakens ties and prompts exploration, enabling broader spread across social, neural, and infection networks; the findings suggest breaking echo chambers by introducing friction in familiar paths and have implications for understanding how habits and viral content propagate, with the work published in Communications Physics.