When logic doesn't matter
Misinformation and extreme ideas tend to spread faster online, and conversations can quickly become more hostile than in face-to-face interactions (Vosoughi, S., et al., Scienceopens in new window, Vol. 359, No. 6380, 2018; Bor, A., & Petersen, M. B., American Political Science Reviewopens in new window, Vol. 116, No. 1, 2022). In a Nature survey, nearly 60% of scientists who had commented publicly about COVID-19 online or in the media said they faced attacks to their credibility, and 15% said they received death threats (Nogrady, B., Natureopens in new window, Vol. 598, 2021). And when trolls team up for a coordinated barrage, “it is truly overwhelming,” Van Bavel said.
At the heart of the problem is our hyper-moralized political culture, said psychologist Peter Ditto, PhD, of the University of California, Irvine. He, Van Bavel, and an interdisciplinary group of other scholars are using the term “political sectarianism” to describe what resembles a religious battle between two warring sects, with each side convinced they are right (Scienceopens in new window, Vol. 370, No. 6516, 2020).
“When a scientist like Jay speaks out, he’s viewed not just as wrong but as sacrilegious and even evil,” Ditto said.
While the attacks surrounding vaccination, climate change, and other scientific findings may seem like a debate about what’s factually correct, the conversation is actually more of a moral one.
“The trolls aren’t saying your sample size isn’t big enough; they’re going after your allegiances,” Ditto said.
That can include harassing a scientist simply because of their ties to academia or because they conduct or comment on research relating to COVID-19. Online trolling also disproportionately affects members of nondominant identity groups, such as women, transgender people, and people of color (Vogels, E. A., “The State of Online Harassmentopens in new window,” Pew Research Center, 2021).
Online attacks often escalate when an influencer—someone like Joe Rogan or Jack Posobiec—goads their supporters to attack a perceived enemy. The psychological principle of “engaged followership” helps explain how people justify attacking others in service of a revered leader or ideology (Haslam, S. A., & Reicher, S. D., Annual Review of Law and Social Scienceopens in new window, Vol. 13, 2017).
“In many cases, online trolls probably feel that what they’re doing is righteous,” said psychologist Gordon Pennycook, PhD, an assistant professor at the University of Regina in Saskatchewan, Canada, who was also harassed on Twitter after backing Van Bavel. “But the problem is that their reality is so miscalibrated that their behavior becomes irrational.”
The collective behavior of online mobs also resembles the dynamics within cults. Cults provide a powerful component of social belonging, which can make extreme ideas feel more palatable, according to the classic theories of social psychologist Leon Festinger, PhD.
“If you have other people helping you rationalize your beliefs and explain away contradictions, it’s easier to cling to these extreme ideas,” said Van Bavel.
Conspiracy theories appear to spread even faster on social media than they do in person, possibly because our cognitive biases function differently online than off, according to research by Mason Youngblood, PhD, a psychologist and researcher at the Max Planck Institute for the Science of Human History in Jena, Germany (Humanities and Social Sciences Communicationsopens in new window, Vol. 7, 2020). For instance, people may be more sensitive to social cues indicating ingroup or outgroup status during online interactions.
Like extreme ideas, extreme behaviors can also spread and escalate quickly online. “Pluralistic ignorance” describes a psychological phenomenon in which people make incorrect assumptions about the beliefs of others and adjust their behavior to align with that perceived group norm. For example, a listener may believe that fellow Joe Rogan Experience fans hold more extreme views and participate in an online attack because they (incorrectly) believe that their peers are enjoying it.
Krotoski and others have shown that this explains some aspects of online trolling, because users notice, amplify, and act on very subtle cues about group status (Lee, S. M., et al., Journal of Computer Information Systemsopens in new window, Vol. 48, No. 3, 2016). In the online space, it’s easier for people to feel united against a common enemy—and to harass that person or group simply because they see others doing so.
“Pluralistic ignorance has an extraordinary amplifying effect in the virtual environment,” Krotoski said. “Without some sort of reality check that we would often experience offline, we will very happily continue to escalate.”
[Related: How to handle an online siege]
Staying safe from trolls
Experts stress that any scientist who discusses science on social media or in the popular press risks having their ideas challenged, their credibility questioned, and even their personal safety threatened.
“By participating in the conversation, it’s almost like you sign an unspoken contract, and you could potentially come under fire,” said Krotoski, who has spent 20 years broadcasting on gaming, psychological research, and other topics and has faced repeated online harassment.
Those who want to share their research on social media can take some steps to protect themselves, starting by creating a plan for how to respond if an attack does occur.
Familiarize yourself with privacy tools native to platforms, including how to mute, block, and report other users. Van Bavel also recommends reviewing any “lists” on Twitter that you’ve been added to, because trolls often curate lists of academics to facilitate harassment. Navigate to the “Lists” tab on Twitter and access “Lists you’re on” by clicking the three dots at the top.
External tools can help further protect against harassers. The Block Partyopens in new window app filters out unwanted posts and mentions from a Twitter feed, and the Twitter Block Chainopens in new window browser extension for Chrome allows users to block trolls and their followers en masse.
Decide ahead of time whether and how you will respond to critics. Some challengers may be interested in a reasoned debate—Pennycook said his replies to skeptics have, at times, led to interesting discussions. Other approaches include choosing not to respond to comments or private messages or having a colleague review your replies before sending.
Next, researchers should know how to recognize when an online mob is forming. One red flag is a sudden surge of comments and messages, especially if people are pushing conspiracy theories or accusing researchers of lying.
If you are targeted, you can wait until the storm passes or address it head-on. Van Bavel chose to temporarily make his Twitter account private and hunker down until the trolls moved on. In some cases, a scholar who feels their findings or comments have been misrepresented may elect to release a video or written statement explaining their position.
“Whatever you do, just don’t feed the trolls” by retaliating at those who are attacking you, Krotoski said. When hundreds or thousands of people are attacking your character and credibility, responding to individual critics in an aggressive or defensive way will usually make things worse.
Researchers who anticipate pushback against a public statement, for instance about vaccination or another hot-button issue, may want to solicit support from colleagues before going public, suggested Heidi Tworek, PhD, an associate professor of public policy and global affairs and history at the University of British Columbia who studies online harassment of health communicators. For example, arrange to have several other scientists publicly validate your comments as soon as they are posted or released.
When the moment comes, set your plan in action and prepare to weather the storm. Though the deluge can feel intense, it typically doesn’t last very long, Pennycook said.
“Trolls are very reactionary and tend to move on quickly. It’s a lot really fast, and then the next thing happens,” he said.
Scientists in the public eye should also periodically audit their online footprint. For a small fee, DeleteMeopens in new window delists personal information, such as addresses and phone numbers, from public databases. TweetDeleteropens in new window and other services can periodically archive social media posts to cut down on fodder for trolls. Various guidesopens in new window also offer detailed instructions for removing personal information from websites frequently used by online attackers.
Other tips include making personal profiles—such as an Instagram account with family photos—private, applying for Twitter’s blue verification badge to minimize the risk of impostor accounts, and working with your institution to remove your contact information from its public directory.
Social and institutional support
While individuals can take steps to enhance their personal safety, experts emphasize that institutions, governments, funding agencies, scientific societies, and media platforms are ultimately responsible for creating safer online spaces.
“A lot of this is about the ecosystem of the online world, and that cannot be solved by individuals,” Tworek said.
In particular, institutions that encourage researchers to engage with public audiences should have systems in place to support those who face abuse when doing so, she said. For example, universities can form a safety office staffed with individuals trained to help academics block harassers on social media, report death threats to the police, or filter through hundreds of abusive emails and voicemails. A centralized office could also expedite the removal of contact information from university websites and connect anyone facing abuse with a counselor.
When conspiracy theorists emailed NYU’s administrative leaders to accuse Van Bavel of unethical behavior and of feeding propaganda to the media, he received a personal message of solidarity from the NYU provost. But he acknowledges that scientists in more precarious employment situations—including academics without tenure or clinicians in private practice who rely on online reviews—may fear serious reputational damage from such an attack.
For that reason, he emphasized the importance of supporting colleagues who find themselves in this situation. Support from those who are less likely to be attacked based on their gender, race, or academic status can be especially powerful.
“If you see another researcher being harassed, please step in or recommend another expert who’s comfortable doing so,” Van Bavel said. “This experience has underscored to me how critical that social support really is.”
