Strange things have been happening in the news lately. Already this year, members of US President Donald Trump's administration have alluded to a 'Bowling Green massacre' and terror attacks in Sweden and Atlanta, Georgia, that never happened.
The misinformation was swiftly corrected, but some historical myths have proved difficult to erase. Since at least 2010, for example, an online community has shared the apparently unshakeable recollection of Nelson Mandela dying in prison in the 1980s, despite the fact that he lived until 2013, leaving prison in 1990 and going on to serve as South Africa's first black president.
Memory is notoriously fallible, but some experts worry that a new phenomenon is emerging. “Memories are shared among groups in novel ways through sites such as Facebook and Instagram, blurring the line between individual and collective memories,” says psychologist Daniel Schacter, who studies memory at Harvard University in Cambridge, Massachusetts. “The development of Internet-based misinformation, such as recently well-publicized fake news sites, has the potential to distort individual and collective memories in disturbing ways.”
Collective memories form the basis of history, and people's understanding of history shapes how they think about the future. The fictitious terrorist attacks, for example, were cited to justify a travel ban on the citizens of seven “countries of concern”. Although history has frequently been interpreted for political ends, psychologists are now investigating the fundamental processes by which collective memories form, to understand what makes them vulnerable to distortion. They show that social networks powerfully shape memory, and that people need little prompting to conform to a majority recollection — even if it is wrong. Not all the findings are gloomy, however. Research is pointing to ways of dislodging false memories or preventing them from forming in the first place.
To combat the influence of fake news, says Micah Edelson, a memory researcher at the University of Zurich in Switzerland, “it's important to understand not only the creation of these sites, but also how people respond to them”.
All together now
Communication shapes memory. Research on pairs of people conversing about the past shows that a speaker can reinforce aspects of an event by selectively repeating them1. That makes sense. Things that get mentioned get remembered — by both speaker and listener. There's a less obvious corollary: related information that goes unmentioned is more likely to fade than unrelated material, an effect known as retrieval-induced forgetting.
These cognitive, individual-level phenomena have been proposed as a mechanism for memory convergence — the process by which two or more people come to agree on what happened. But in the past few years, clues have emerged that group-level forces influence convergence, too. In 2015, psychologists Alin Coman at Princeton University in New Jersey and William Hirst of the New School for Social Research in New York City reported that a person experiences more induced forgetting when listening to someone in their own social group — a student at the same university, for example — than if they see that person as an outsider2. That is, memory convergence is more likely to occur within social groups than between them — an important finding in light of survey data suggesting that 62% of US adults get their news from social media, where group membership is often obvious and reinforced3.
Groups can also distort memories. In 2011, Edelson, then at the Weizmann Institute of Science in Rehovot, Israel, showed 30 volunteers a documentary. They watched the film in groups of five and, a few days later, answered questions about it individually. One week after the viewing session, participants answered questions again — but only after seeing answers that members of their group had supposedly given. When most of the fabricated responses were false, participants conformed to the same false answer about 70% of the time — despite having initially responded correctly. But when they learnt that the answers had been generated randomly, the participants reversed their incorrect answers only about 60% of the time4. “We found that processes that happen during initial exposure to erroneous information make it more difficult to correct such influences later,” says Edelson.
Studying those processes as they happen — as collective memories are shaped through conversation — has been difficult to do in large groups. Five years ago, monitoring communication in groups of ten or more would have required several rooms for private conversations, many research assistants and lots of time. Now, multiple participants can interact digitally in real time. Coman's group has developed a software platform that can track exchanges between volunteers in a series of timed chats. “It takes one research assistant 20 minutes and one lab room,” Coman says.
The researchers investigated two scenarios — one in which the group formed two sub-clusters, with almost all conversations taking place within the sub-clusters, and one in which it formed one large cluster (see 'Hello operator'). Although people in the single cluster agreed on the same set of information, says Coman, those in the two sub-clusters generally converged on different 'facts' about the fictional volunteers5.
In the lab, Coman can manipulate social networks and look at the memories that form. His comparison of the two scenarios revealed the importance of 'weak links' in information propagation. These are links between, rather than within, networks — acquaintances, say, rather than friends — and they help to synchronize the versions held by separate networks. “They are probably what drives the formation of community-wide collective memories,” he says.
One function of those weak links might be to remind people of information expunged through the processes of memory convergence. But timing is important. In unpublished work, Coman has shown that information introduced by a weak link is much more likely to shape the network's memory if it is introduced before its members talk among themselves. Once a network agrees on what happened, collective memory becomes relatively resistant to competing information.