An information cascade occurs when people make decisions sequentially rather than simultaneously. Later people watch the actions of earlier people, and base their actions on the earlier ones. An example would be someone who searches for a new song by looking for the most popular songs. The cascade can be based on little information and people could ignore new information. This cascade can easily be based on faulty information, and the credibility of crowds could quickly surpass that provided by traditional experts and authorities. Yet, cascades are fragile and can be overturned by people with more compelling information causing a counter-cascade.