Before June 1, 2013 Turkey’s ruling political party and its leader seemed invincible. They were regarded as the architects of a decade-long economic boom and their public support seemed unshakable. This image shattered in less than a week. They were suddenly described as incompetent and backward with an uncertain political future.

On Friday May 31st, a small group gathered at a park in central Istanbul to protest against government plans to replace the park with a housing complex and a shopping mall. The police attacked the group with tear gas, pepper spray, and water cannons and cleared the park of protesters. The police brutality was not new but the subsequent public reaction was unprecedented. Tens of thousands marched to the park despite continued police violence. On Saturday June 1st, police forces had to retreat and the people occupied Gezi Park and the adjacent Taksim Square. The uprising quickly spread throughout the country and turned into a revolt against the ruling party and its leader’s autocratic style and Islamic agenda.

Twitter became the principal communication tool for the protesters; a diverse and previously politically inactive people self-mobilized; and the uprising caught most observers by surprise. These three features may be interrelated. There is an emerging understanding of the relationship between connectedness and collective decisions. As one would expect, when people are better connected, they tend to unite around popular decisions. But research also suggests that social connection — fostered by Twitter, say — also makes crowds fundamentally less predictable. With social media connecting people to an unprecedented degree, it is possible that the sudden emergence of unexpected collective action will be a defining feature of this era.

Social media became the primary means of communication partly because the government-controlled traditional media was not covering the events. People shared information about what was happening at protest sites as well as their views on the events. The government blocked transportation routes to Taksim Square, so people often covered long distances on foot to arrive there. People constantly tweeted their gathering sites and current locations to organize their march. When volunteer doctors arrived at Taksim Square and surrounding areas to help injured protesters, the doctors tweeted their locations and cell phone numbers. Voluntary lawyers did the same when they were headed to police stations to defend detained protesters.

Twitter was also used in the 2011 Egyptian Revolution, but there are differences. During the Egyptian Revolution, the tweeting was mainly outside Egypt and it was used to inform the outside world. In Turkey’s case, tweets came from within Turkey and were used to facilitate communication between protesters. Also the sheer volume of tweets makes Turkey’s case unique. Turkey’s prime minister was quick to realize how critical social media’s role was. During a June 2nd interview, he called social media a “menace to society.” Shortly after his comments, dozens of Twitter users were arrested.

Matthew Salganik, Peter Dodds, and Duncan Watts conducted large-scale experiments to investigate the effect of the strength of social influence on collective action. People were given a list of previously unknown songs from unknown bands. They listened to the songs and downloaded them if they wanted to. In the independent condition, people did not see other people’s choices. In the social influence condition, people saw how many times each song had been downloaded by others. The collective outcome in the social influence condition was more unequal. That is, popular choices were much more popular under social influence. When the researchers increased the strength of social influence by displaying the songs in a table ordered by popularity, the collective outcome became even more unequal.

After the events around Gezi Park unfolded, there were attempts at explaining why they happened. Events seem inevitable retrospectively. But the truth is most people did not know that the uprising was coming. But was it knowable? In other words, could a sophisticated observer accurately predict the events? Salganik, Dodds, and Watts looked at the collective outcomes in eight different “worlds.” That is, eight separate groups of people downloaded songs under social influence. The collective outcomes in different “worlds” were different. Even though people in different “worlds” were indistinguishable and they did the same task under the same conditions, the collective outcomes were different. When the researchers increased the strength of social influence, the collective outcome became even more unpredictable. That is, the difference between the popularities of a given song in different “worlds” increased as the strength of social influence increased. Apparently, in collective decisions there is an inherent unpredictability that cannot be resolved by carefully examining the initial conditions and decision makers. Stronger social influence results in more unpredictability.

Around the world, people are getting better connected. Despite censorship, Chinese people are well connected through social media and this is already shaping collective behavior. The number of internet users in Iran has been increasing very rapidly. People in more developed countries are already well connected and those countries are not immune from social unrest. People of Spain, Italy, and Greece have been experiencing dwindling incomes and high youth unemployment. In the U.S. meanwhile, the class structure became so rigid that rising from humble origins is now harder in the U.S. than in many other countries. Of course, that kind of widespread social unrest would never happen here, one might think. Yet the recent uprising has been led by Turkey’s youth, which has not shown any significant political activism since 1970s. In fact, they were frequently described as a generation that shows insufficient interest in politics. Given the unpredictable nature of collective action, we should all expect to be surprised by crowds.

Are you a scientist who specializes in neuroscience, cognitive science, or psychology? And have you read a recent peer-reviewed paper that you would like to write about? Please send suggestions to Mind Matters editor Gareth Cook, a Pulitzer prize-winning journalist and regular contributor to He can be reached at garethideas AT or Twitter @garethideas.