Social Media and Morality.

Social media posts that contain moral emotional language are more likely to be shared than those that do not contain moral emotional language.

“social media posts that contain moral emotional language are more likely to be shared on social media than those that do not contain moral emotional language (Brady et al. 2017)”

• cites Brady et al. 2017 sourcePartly supported

  The Results report that moral-emotional language predicted higher retweet rates across tweets about gun control, same-sex marriage, and climate change. The Discussion likewise describes increased transmission of political tweets, but the cited study does not establish the claim for social media posts generally. (LLM, conf=0.83)

  Experimental — accuracy not yet measured

The absence of nonverbal cues in text-based social media may exaggerate the perceived intensity of users' outrage.

“the absence of nonverbal cues that are essential in face-to-face communication may exaggerate the perceived intensity of users' outrage (Lieberman & Schroeder 2020)”

• cites Lieberman & Schroeder 2020 sourcePartly supported

  The “Fewer nonverbal cues online” section (page not provided) says text-based platforms lack nonverbal cues and that reading opinions rather than hearing them can impair judgments of a communicator’s feelings. The “Greater anonymity online” section separately discusses moral outrage on social media. Neither passage says that missing cues exaggerate the perceived intensity of users’ outrage. (LLM, conf=0.84)

  Experimental — accuracy not yet measured

People perceive online discussions to be far more hostile than offline discussions.

“people nevertheless perceive online discussions to be far more hostile (Bor & Petersen 2022)”

• cites Bor & Petersen 2022 sourceSupported

Some people who are exposed to opposing partisan accounts on social media experience a backfire effect, becoming more entrenched in their own beliefs.

“one online experiment found that some people who are exposed to opposing partisan accounts on social media actually experience a backfire effect, such that they become more entrenched in their own beliefs (Bail et al. 2018)”

• cites Bail et al. 2018 sourceSupported

Mentioning an out-group member is the strongest predictor of social media engagement out of a variety of predictors measured in an analysis of Facebook and Twitter posts (N = 2,730,215) by politicians and partisan news media sources.

“mentioning an out-group member is the strongest predictor of social media engagement out of a variety of predictors measured in an analysis of Facebook and Twitter posts (N = 2,730,215) by politicians and partisan news media sources (Rathje et al. 2021)”

• cites Rathje et al. 2021 sourceSupported

Aggressive individuals appear to act equally aggressive in both offline and online contexts.

“aggressive individuals appear to act equally aggressive in both offline and online contexts (Bor & Petersen 2022)”

• cites Bor & Petersen 2022 sourceSupported

People who engage in aggressive and hostile behaviors in the offline world are also likely to do so in the online world, where their actions are accelerated by the structure of the Internet.

“People who engage in aggressive and hostile behaviors in the offline world are also likely to do so in the online world, where their actions are accelerated by the structure of the Internet (Bor & Petersen 2022;Kowalski et al. 2014Kowalski et al. , 2019))”

• cites Bor & Petersen 2022 sourcePartly supported

  The Conclusions and General Discussion say that individuals motivated by status and dominance engage in hostility in both online and offline political discussions. Study 8 and the Conclusions attribute the greater apparent online hostility to the Internet’s connectivity, which makes hostile actions more visible and gives them greater reach. They do not say that the Internet accelerates the actions themselves, and the findings concern political hostility rather than aggressive behavior generally. (LLM, conf=0.86)

  Experimental — accuracy not yet measured
• cites Kowalski et al. 2014 sourceNot checked — no full text of the source
• cites Kowalski et al. 2019 sourceNot checked — no full text of the source

Feelings of marginalization coupled with dominance orientation make people more likely to engage in hostility on social media.

“Studies suggest that feelings of marginalization coupled with dominance orientation make people more likely to engage in hostility on social media (Bor & Petersen 2022, Petersen et al. 2023)”

• cites Bor & Petersen 2022 sourcePartly supported

  The Results (page not provided) find that hostility increases with status-driven risk taking in online political discussions. The Conclusions and General Discussion (page not provided) also describe status and dominance as motives for hostility. Neither passage supports the added claim that feelings of marginalization are coupled with dominance orientation. (LLM, conf=0.88)

  Experimental — accuracy not yet measured
• cites Petersen et al. 2023 sourceNot checked — no full text of the source

Because there is a large volume of people to connect with online, there are significantly more extreme voices to affiliate with than in prior social groups limited by geography.

“Because there is such a large volume of people to connect with in the online context, there are significantly more extreme voices to affiliate with than there were in prior social groups that were limited by geography (Petersen et al. 2020, Törnberg 2022).”

• cites Petersen et al. 2020 sourceNot checked — no full text of the source
• cites Törnberg 2022 sourcePartly supported

  The Discussion (page not provided) says digital media connect people outside their local social bubbles, supporting the comparison with geographically limited groups. It does not say that online networks offer significantly more extreme voices to affiliate with; the paper instead explains polarization through nonlocal interaction and partisan sorting. (LLM, conf=0.85)

  Experimental — accuracy not yet measured

Part of the trend of 3G Internet access increasing support for populist parties might be explained by the Internet informing users about government corruption.

“Part of this trend might be explained by the Internet's informing users about government corruption (Zhuravskaya et al. 2020).”

• cites Zhuravskaya et al. 2020 sourcePartly supported

  Section 2.1 reports that expanding 3G coverage benefited populist opposition parties in Europe. It also says mobile Internet exposed government corruption and suggests that informing voters explains part of its effect on government approval in a global analysis. The review does not establish that corruption information explains the specifically European populist-vote trend. (LLM, conf=0.82)

  Experimental — accuracy not yet measured

Partisan polarization is the primary psychological motivation behind fake news sharing.

“One study of fake news sharing on Twitter found that partisan polarization is the primary psychological motivation behind fake news sharing”

• cites Osmundsen et al. 2021 sourceSupported

People tend to overestimate the degree of polarization between political groups.

“People also tend to overestimate the degree of polarization (Levendusky & Malhotra 2016, Westfall et al. 2015) and animosity (Mernyk et al. 2022, Ruggeri et al. 2021) between political groups, which may be exacerbated by the abundance of divisive content they see online.”

• cites Levendusky & Malhotra 2016 sourceSupported
• cites Westfall et al. 2015 sourceNot checked — no full text of the source

Conspiracy believers may self-select into online communities that share and reinforce their beliefs.

“This may lead conspiracy believers to self-select into online communities that in turn share and reinforce their beliefs (Robertson et al. 2022).”

• cites Robertson et al. 2022 sourceSupported

People tend to overestimate animosity between political groups.

“People also tend to overestimate the degree of polarization (Levendusky & Malhotra 2016, Westfall et al. 2015) and animosity (Mernyk et al. 2022, Ruggeri et al. 2021) between political groups, which may be exacerbated by the abundance of divisive content they see online.”

• cites Mernyk et al. 2022 sourcePartly supported

  The Results (Studies 1 and 2) show that Democrats and Republicans overestimated rival partisans’ support for violence. That supports an overestimation of a specific hostile attitude, but the paper does not establish that people generally overestimate animosity between political groups. (LLM, conf=0.82)

  Experimental — accuracy not yet measured
• cites Ruggeri et al. 2021 sourceSupported

Another potential consequence of the spread of moral and emotional content online is increased polarization.

“Another potential consequence of the spread of moral and emotional content online is increased polarization (Iyengar et al. 2019).”

• cites Iyengar et al. 2019 sourcePartly supported

  The ‘Origins and Causes of Affective Polarization’ section (pp. 6–7) describes hostile partisan content and says social networks can amplify polarizing messages. It supports online content as a potential contributor to polarization, but does not specifically identify the spread of moral and emotional content online as the contributor. (LLM, conf=0.83)

  Experimental — accuracy not yet measured

True news is more likely to generate engagement than false news on Reddit.

“Additionally, true news is more likely to generate engagement than false news on Reddit (Bond & Garrett 2023).”

• cites Bond & Garrett 2023 sourcePartly supported

  The Discussion reports lower engagement for fact-checked Reddit posts whose information was false than for those whose information was true. The claim omits that the study examined posts eliciting fact-checking comments, so it extends the finding to Reddit news generally. (LLM, conf=0.86)

  Experimental — accuracy not yet measured

Misinformation spread on Twitter is linked to lower county-level vaccine uptake.

“Other research suggests that misinformation spread on Twitter is linked to lower county-level vaccine uptake (Pierri et al. 2022), and people who got their news primarily from”

• cites Pierri et al. 2022 sourcePartly supported

  The Results report that Twitter misinformation was significantly correlated with county-level vaccine hesitancy. The Methods distinguish that outcome from vaccination uptake, which the study measured only at the state level; the cited paper therefore does not establish lower county-level uptake. (LLM, conf=0.91)

  Experimental — accuracy not yet measured

Americans with anti-vaccine and pro-vaccine attitudes clustered into distinct echo chambers on Twitter.

“Americans with anti-vaccine and pro-vaccine attitudes clustered into distinct echo chambers on Twitter (Rathje et al. 2022).”

• cites Rathje et al. 2022 sourceSupported

There is a causal link between exposure to vaccine misinformation and reduced intentions to get vaccinated.

“Other research has found a causal link between exposure to vaccine misinformation and reduced intentions to get vaccinated (Loomba et al. 2021).”

• cites Loomba et al. 2021 sourceSupported

Many studies have found that misinformation and moral outrage are often closely intertwined.

“Many studies have found that misinformation and moral outrage are often closely intertwined (Marie et al. 2023, Vosoughi et al. 2018).”

• cites Marie et al. 2023 sourcePartly supported

  The Discussion of Experiments 1–4 reports that moralizing a political issue amplified participants’ willingness to share politically congruent news, including fake news. That supports a link between moralization and misinformation sharing, but the paper does not establish a link specifically with moral outrage or show that the two are often intertwined. (LLM, conf=0.83)

  Experimental — accuracy not yet measured
• cites Vosoughi et al. 2018 sourceNot checked — no full text of the source

Polarized rhetoric about the pandemic on Twitter was highly correlated with polarized beliefs about the risks of the pandemic several days later.

“For instance, polarized rhetoric about the pandemic on Twitter was highly correlated with polarized beliefs about the risks of the pandemic several days later (Simchon et al. 2022).”

• cites Simchon et al. 2022 sourcePartly supported

  The Dictionary validation section reports a strong correlation (r = 0.67) between polarized COVID-19 language on Twitter and partisan differences in pandemic concern at an eight-day offset. It does not clearly support the claimed direction of that offset: the passage specifies poll responses at t0 and Twitter language at t8, rather than beliefs occurring several days after the tweets. (LLM, conf=0.82)

  Experimental — accuracy not yet measured

Early in the COVID-19 pandemic, attitudes and behavior related to the pandemic were especially polarized, with Republicans and Democrats in the United States having strikingly different attitudes about public health behaviors and the COVID-19 vaccine.

“Early in the COVID-19 pandemic, attitudes and behavior related to the pandemic were especially polarized, with Republicans and Democrats in the United States having strikingly different attitudes about public health behaviors and the COVID-19 vaccine (Gollwitzer et al. 2020, Van Bavel et al. 2023).”

• cites Gollwitzer et al. 2020 sourcePartly supported

  Split into 2 atoms — 1 confirmed, 1 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

  Experimental — accuracy not yet measured
• cites Van Bavel et al. 2023Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Because social media platforms allow information to spread with less friction than in traditional news outlets, misinformation and conspiracy theories are able to go rapidly viral.

“Because social media platforms allow information to spread with less friction than in traditional news outlets, such as newspapers or television news, misinformation and conspiracy theories are able to go rapidly viral (Robertson et al. 2022, Van Bavel et al. 2021a, van der Linden et al. 2021).”

• cites Robertson et al. 2022 sourcePartly supported

  The Introduction says internet and social media let conspiracy theories spread through large communities, and New Directions says they may go viral online. The paper does not compare social media with newspapers or television, attribute the spread to lower friction, or establish that misinformation generally goes rapidly viral. (LLM, conf=0.86)

  Experimental — accuracy not yet measured
• cites Van Bavel et al. 2021aNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites van der Linden et al. 2021 sourceNot checked — no full text of the source

Deleting Facebook during genocide remembrance week in Bosnia increased ethnic polarization, especially if people had particularly homogeneous offline social networks.

“For example, deleting Facebook during genocide remembrance week in Bosnia actually increased ethnic polarization, especially if people had particularly homogeneous offline social networks (Asimovic et al. 2021).”

• cites Asimovic et al. 2021 sourcePartly supported

  The Abstract and Research Design report that deactivating Facebook during the Srebrenica genocide commemoration week led to more negative attitudes toward ethnic outgroups. The Discussion finds a stronger effect among residents of ethnically homogeneous towns, but uses town composition as a proxy for offline networks; that analysis was not preregistered. The claim presents homogeneous offline networks as an established individual-level condition rather than a tentative inference from where participants lived. Page numbers were not provided in our copy. (LLM, conf=0.85)

  Experimental — accuracy not yet measured

The large majority of published research has focused on North American and European issues and samples.

“The large majority of published research has focused on North American and European issues and samples.”

• cites Kubin & von Sikorski 2021 sourceCould not verify

  The Results section lists a “Country of sample” heading, and the appendix codebook lists “Country of sample” and “Political topic,” but the available passages give no findings under those headings. Searches for North American and European issues and samples, and for the distribution of sample countries, found no passage establishing where the large majority of research was focused. (LLM, conf=0.70)

  Experimental — accuracy not yet measured

Peer influence may be effective in shaping norms in social media environments when the peers are credible members of one's in-group.

“This suggests that peer influence may be effective in shaping norms in social media environments when the peers are credible members of one's in-group (Paluck & Green 2009, Paluck et al. 2021).”

• cites Paluck & Green 2009 sourceNot checked — no full text of the source
• cites Paluck et al. 2021 sourceSupported

86% of the research on social media and polarization is from the United States.

“For instance, 86% of the research on social media and polarization is from the United States (Kubin & von Sikorski 2021).”

• cites Kubin & von Sikorski 2021 sourceCould not verify

  The available Results passage lists a “Country of sample” section but contains no findings or percentages. Searches for “86%” and the share of studies using U.S. samples returned no passage establishing the claimed figure; the available copy appears to contain headings and links rather than the relevant body text. (LLM, conf=0.72)

  Experimental — accuracy not yet measured

Identity- and empathy-based bystander interventions on social media can effectively reduce incivility, hate speech, and misinformation.

“For instance, identity-and empathy-based bystander interventions on social media can effectively reduce incivility (Munger 2020), hate speech (Hangartner et al. 2021, Munger 2017, Siegel & Badaan 2020), and misinformation (Pretus et al. 2022).”

• cites Munger 2020 sourceNot checked — no full text of the source
• cites Hangartner et al. 2021 sourceSupported
• cites Munger 2017 sourceNot checked — no full text of the source

People across the political spectrum believe that misinformation and hate speech should be restricted based on severity, regardless of the target or topic.

“People across the political spectrum believe that misinformation and hate speech should be restricted based on severity, regardless of the target or topic (Kozyreva et al. 2023, Rasmussen 2022, Rathje et al. 2023).”

• cites Kozyreva et al. 2023 sourcePartly supported

  The Discussion reports that US respondents were more willing to restrict misinformation as the severity of its consequences increased, and that the account holder’s characteristics had little effect. But it explicitly says the misinformation topic mattered, with climate-change denial acted on less often than Holocaust or election denial. This paper studies misinformation moderation, not hate-speech restrictions or whether the target of hate speech affects decisions. (LLM, conf=0.90)

  Experimental — accuracy not yet measured
• cites Rasmussen 2022Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Rathje et al. 2023Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

People on the political left are more supportive of regulating social media platforms.

“In general, people on the political left are more supportive of regulating social media platforms (Kozyreva et al. 2023, Rasmussen 2022), reflecting liberals' general tendency to be more supportive, and conservatives' tendency to more opposing, of regulatory policies (e.g., Grossman & Hopkins 2016).”

• cites Kozyreva et al. 2023 sourcePartly supported

  The Discussion reports that Democrats preferred preventing dangerous falsehoods across four content-moderation scenarios, while Republicans preferred free speech and imposed fewer restrictions. That supports a partisan difference in attitudes toward moderating online misinformation, but the paper limits its conclusions to US respondents and does not establish a general preference for regulating social media platforms across issues. (LLM, conf=0.84)

  Experimental — accuracy not yet measured
• cites Rasmussen 2022Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Relational mobility afforded by social media ensures that even those who burn bridges with their original social groups can find new groups that agree with their extreme opinions.

“The relational mobility afforded by social media ensures that even those who burn bridges with their original social groups can find new groups that agree with their extreme opinions.”

The design and incentives of social media accelerate moral content.

“The design and incentives of social media accelerate moral content.”

Social media potentially amplifies positive aspects of morality, such as social support, prosociality, and collective action.

“while also potentially amplifying more constructive facets of morality, such as social support, prosociality, and collective action”

• Also stated in: § INTRODUCTION · plain

  “as well as some positive aspects of morality, such as social support, prosociality, and collective action.”

Moral outrage can encourage an active response, such as protests for social justice.

“moral outrage can encourage an active response, such as protests for social justice (Hutcherson & Gross 2011, Jost et al. 2018, Spring et al. 2018, van Zomeren et al. 2004)”

• cites Hutcherson & Gross 2011 sourceNot checked — no full text of the source
• cites Jost et al. 2018 sourceNot checked — no full text of the source
• cites Spring et al. 2018 sourceNot checked — no full text of the source

Social media makes moral behavior, such as punishment, less costly.

“social media makes moral behavior (such as punishment) less costly”

• Also stated in: § Abstract · plain

  “social media makes moral behavior (such as punishment) less costly.”

One means by which individuals and groups can capture attention and drive engagement on social media is by sharing morally and emotionally evocative content.

“One means by which individuals and groups can capture attention and drive engagement on these platforms is by sharing morally and emotionally evocative content”

• Also stated in: § INTRODUCTION · plain

  “One means by which individuals can capture attention and drive engagement is by sharing morally and emotionally evocative content (see Brady et al. 2020a,b).”

• cites Brady et al. 2020a,bNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

People tend to misinterpret social media posts as expressing greater outrage than the author actually feels.

“people tend to misinterpret social media posts as expressing greater outrage than the author actually feels (Brady et al. 2023)”

• cites Brady et al. 2023 sourceNot checked — no full text of the source

Social media often acts as an accelerant for existing moral dynamics, amplifying negative facets of morality such as outrage, harassment, status seeking, and intergroup conflict.

“social media often acts as an accelerant for existing moral dynamics, amplifying outrage, status seeking, and intergroup conflict”

• Also stated in: § INTRODUCTION · hedged

  “Our review of the literature suggests that social media often acts like an accelerant for existing moral dynamics-amplifying negative facets of morality such as outrage, harassment, status seeking, and intergroup conflict”

Social media offers high-visibility and low-cost forms of engagement, such as comments and likes, which make it easy for users to punish, shame, and pile on transgressors.

“social media offers high-visibility and low-cost forms of engagement, such as comments and likes, which make it easy for users to punish, shame, and pile on transgressors”

Social media algorithms further accelerate the cycle of outrage by promoting related ideological content.

“Social media algorithms further accelerate this cycle of outrage by promoting related ideological content.”

Moral content often goes viral on social media.

“Moral content often goes viral on social media”

Feedback from low-cost engagement can serve as reinforcement learning that further fuels outrage.

“This feedback can then serve as a type of reinforcement learning that further fuels outrage (Brady et al. 2021)”

• cites Brady et al. 2021Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Anonymity of social media may embolden some people to express views and emotions they may not disclose in face-to-face interactions.

“The anonymity of social media may also embolden some people to express views and emotions they may not disclose in face-to-face interactions (Nitschinsk et al. 2022, Zimmerman & Ybarra 2016)”

• cites Nitschinsk et al. 2022 sourceNot checked — no full text of the source
• cites Zimmerman & Ybarra 2016Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Social media reproduces, and often exaggerates, the social behavior and intergroup dynamics that emerge from group identities, moral sensibilities, and political polarization.

“social media reproduces-and often exaggerates-the social behavior and intergroup dynamics that emerge from our group identities, moral sensibilities, and political polarization”

Social media can amplify norm following and enforcement at a speed and scale difficult to implement in the real world.

“social media can amplify norm following and enforcement at a speed and scale that would be difficult to implement in the real world”

The larger the number of people who publicly criticize a moral transgression, the more likely third parties are to empathize with the original perpetrator.

“The larger the number of people who publicly criticize a moral transgression, the more likely third parties are to empathize with the original perpetrator of the transgression; this is known as”

• cites Sawaoka & Monin 2018 sourceNot checked — no full text of the source

Each moral-emotional word added to a social media post is associated with 12% more retweets.

“each moral-emotional word added to a social media post is associated with 12% more retweets (Brady & Van Bavel 2021a)”

• cites Brady & Van Bavel 2021aNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Moral content captures attention.

“moral content captures attention (Brady et al. 2020b)”

• cites Brady et al. 2020bNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

In discussions with like-minded individuals, people can be persuaded to adopt more extreme ideological positions, leaving an interaction with more polarized views than they initially had.

“In discussions with like-minded individuals, people can be persuaded to adopt more extreme ideological positions, leaving an interaction with more polarized views than they initially had (Mackie & Cooper 1984, Moscovici & Zavalloni 1969).”

• cites Mackie & Cooper 1984 sourceNot checked — no full text of the source
• cites Moscovici & Zavalloni 1969 sourceNot checked — no full text of the source

Extremists' participation online may have a polarizing effect on people's more moderate ideological views.

“Extremists' participation online may have a polarizing effect on peoples' more moderate ideological views.”

Social media might cause people to encounter more divisive and polarizing content than they actually want to see.

“Social media might also cause people to encounter more divisive and polarizing content than they actually want to see.”

The abundance of divisive content people see online may exacerbate overestimation of polarization and animosity between political groups.

“People also tend to overestimate the degree of polarization (Levendusky & Malhotra 2016, Westfall et al. 2015) and animosity (Mernyk et al. 2022, Ruggeri et al. 2021) between political groups, which may be exacerbated by the abundance of divisive content they see online.”

Extremists' contribution on social media may inaccurately depict political groups as more polarized than they truly are (false polarization).

“In other words, extremists' contribution on social media may inaccurately depict political groups as more polarized than they truly are (known as false polarization), yet their participation may lead to real ramifications in which they convince others of their extreme views.”

The push toward extremist views may result in social media acting as a catalyst for conspiratorial thinking.

“This push toward extremist views may also result in social media acting as a catalyst for conspiratorial thinking.”

Massive rallies that emerge on the streets overnight often fizzle just as quickly.

“The result is massive rallies that emerge on the streets overnight but often fizzle just as quickly (Fisher 2022a).”

• cites Fisher 2022aNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Psychopathy is the strongest and most consistent predictor of antisocial online behaviors, followed by Machiavellianism and everyday sadism.

“A meta-analysis of antisocial online behaviors found that psychopathy was the strongest and most consistent predictor of antisocial online behaviors, followed by Machiavellianism and everyday sadism (Moor & Anderson 2019)”

• cites Moor & Anderson 2019 sourceNot checked — no full text of the source

Political extremists, both liberal and conservative, show higher levels of outrage on social media compared to politically moderate users.

“Political extremists, both liberal and conservative, show higher levels of outrage on social media compared to politically moderate users (Brady et al. 2021).”

• cites Brady et al. 2021Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Status seeking via outrage expression is effective because people often equate moral outrage expressions with the speaker's own morality.

“Status seeking via outrage expression is effective because people often equate moral outrage expressions with the speaker's own morality (Skitka 2010, Tosi & Warmke 2016)”

• cites Skitka 2010 sourceNot checked — no full text of the source
• cites Tosi & Warmke 2016 sourceNot checked — no full text of the source

Lack of gatekeepers can facilitate organizing and collective action.

“This lack of gatekeepers can facilitate organizing and collective action (Spring et al. 2018).”

• cites Spring et al. 2018 sourceNot checked — no full text of the source

Heavy social media users are prone to false consensus effects, overestimating how widely held their views are.

“In fact, heavy social media users are prone to false consensus effects, meaning they overestimate how widely held their views are (Bunker & Varnum 2021).”

• cites Bunker & Varnum 2021 sourceNot checked — no full text of the source

Social media might be making nonviolent protests less effective because they are easier to organize without leadership, which makes them less enduring.

“For example, social media might be making nonviolent protests less effective because they are easier to organize without leadership, which makes them less enduring.”

• cites Fisher 2022aNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Social media use is associated with greater conspiratorial beliefs among people predisposed to conspiratorial thinking.

“Social media use is associated with greater conspiratorial beliefs among people who are predisposed to conspiratorial thinking (Enders et al. 2023).”

• cites Enders et al. 2023Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Extremists' participation may lead to real ramifications in which they convince others of their extreme views.

“In other words, extremists' contribution on social media may inaccurately depict political groups as more polarized than they truly are (known as false polarization), yet their participation may lead to real ramifications in which they convince others of their extreme views.”

Open access to online platforms can be beneficial, enabling marginalized voices to face lower barriers to disseminate moral messages and find social support.

“Open access to online platforms can be beneficial, enabling marginalized voices to face lower barriers to disseminate moral messages and find social support (Schmitz et al. 2022).”

• cites Schmitz et al. 2022Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Anyone has the opportunity to create a social media account and post content that rapidly goes viral even if that content consists of baseless conspiracy theories or misinformation.

“Anyone has the opportunity to create a social media account and post content that rapidly goes viral even if that content consists of baseless conspiracy theories or misinformation (Vosoughi et al. 2018).”

• cites Vosoughi et al. 2018 sourceNot checked — no full text of the source

It is extremely difficult to evaluate the net moral impact of social media on society.

“This makes it extremely difficult to evaluate the net moral impact of social media on society (see Brady & Crockett 2019).”

• cites Brady & Crockett 2019Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Moral discourse online is dominated by ideological extremists.

“Moral discourse online is dominated by ideological extremists.”

• cites Brady et al. 2021Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Even though people engage with divisive content on social media, this behavior does not align with people's stated preferences.

“In other words, even though people engage with divisive content on social media-perhaps because it captures attention and drives engagement-this behavior does not align with people's stated preferences (see Figure 3).”

• AssumptionRests on an assumption: People engage with divisive content on social media because it captures attention and drives engagement.

Extremists may incite others to become more ideologically polarized.

“Extremists tend to be dogmatic and to present their beliefs as facts (Harris & Van Bavel 2020, Toner et al. 2013), and they may incite others to become more ideologically polarized.”

The amount of moral convergence in online groups predicts radicalization.

“Moreover, the amount of moral convergence in online groups (e.g., people in a group sharing the same moral views) predicts radicalization (Atari et al. 2022), and, as we mentioned above, tweets containing moral-emotional language are more likely to receive hate speech in the replies (Solovev & Pröllochs 2023).”

• cites Atari et al. 2022 sourceNot checked — no full text of the source

People who got their news primarily from Facebook were even more vaccine hesitant than those who got their news primarily from Fox News.

“Facebook were even more vaccine hesitant than those who got their news primarily from Fox News (Lazer et al. 2021).”

• cites Lazer et al. 2021 sourceNot checked — no full text of the source

Algorithms seem to prioritize the spread of information that is emotional, is polarizing, or captures our attention.

“Algorithms seem to prioritize the spread of information that is emotional, is polarizing, or captures our attention (Brady et al. 2020a,b).”

• cites Brady et al. 2020a,bNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

A full 23% of political images on Facebook contain misinformation.

“For instance, a full 23% of political images on Facebook contain misinformation (Yang et al. 2023).”

• cites Yang et al. 2023Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

COVID-19 rumors that contained moral-emotional language were more likely to spread widely on Twitter.

“One analysis found that COVID-19 rumors that contained moral-emotional language were more likely to spread widely on Twitter (Solovev & Pröllochs 2022).”

• cites Solovev & Pröllochs 2022Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Moralized rhetoric online may spill over into real-world violence.

“Moreover, moralized rhetoric online may spill over into real-world violence (Fisher 2022b, Mooijman et al. 2018).”

• cites Fisher 2022bNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Mooijman et al. 2018 sourceNot checked — no full text of the source

Social media may have different effects on polarization in non-US contexts.

“However, social media may have different effects on polarization in non-US contexts.”

Each additional moral word is associated with a 11-16% increase in eliciting hate speech from readers.

“Moreover, using moral-emotional rhetoric can trigger hate speech among recipients: Each additional moral word is associated with a 11-16% increase in eliciting hate speech from readers (Solovev & Pröllochs 2023).”

• cites Solovev & Pröllochs 2023Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

The increased spread of moralized content online may further fuel polarization and intergroup conflict.

“The increased spread of moralized content online may further fuel polarization and intergroup conflict.”

Social media may be responsible in part for polarization around public health beliefs.

“Social media may be responsible in part for this polarization around public health beliefs.”

Facebook was implicated in an outbreak of ethnic cleansing in Myanmar against the Rohingya.

“For instance, Facebook was implicated in an outbreak of ethnic cleansing in Myanmar against the Rohingya (Fisher 2022b).”

• cites Fisher 2022bNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Algorithms that promote outrage may prioritize promoting false information and conspiracy theories.

“Thus, algorithms that promote outrage may prioritize promoting false information and conspiracy theories.”

Fact-checked false stories spread faster and farther than fact-checked true stories on Twitter.

“One study found that fact-checked false stories spread faster and farther than factchecked true stories on Twitter (Vosoughi et al. 2018).”

• cites Vosoughi et al. 2018 sourceNot checked — no full text of the source

Misinformation containing moral outrage was more likely to spread than misinformation that did not contain moral outrage on Facebook and Twitter.

“Another analysis of Facebook and Twitter posts found that the misinformation containing moral outrage (as detected by a moral outrage machine-learning classifier) was more likely to spread than misinformation that did not contain moral outrage (McLoughlin et al. 2021).”

• cites McLoughlin et al. 2021Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

New research suggests there is more consensus in public opinion on regulating social media than previously thought.

“However, new research suggests there is more consensus in public opinion on regulating social media than previously thought.”

Social media seems to amplify exposure to moral content and action.

“Social media seems to amplify exposure to moral content and action.”

These false beliefs help explain why social media can fuel moral conflicts.

“These false beliefs help explain why social media can fuel (moral) conflicts, as most people are more moderate than people think (Levendusky & Malhotra 2016) and consume news from the same sources on social media (Guess 2021), while those who engage on social media are the most engaged partisans (Settle 2018).”

9 out of 10 countries with the highest numbers of Facebook users are located in the Global South.

“This is despite the fact that 9 out of 10 countries with the highest numbers of Facebook users are located in the Global South (Ghai et al. 2022).”

• cites Ghai et al. 2022 sourceNot checked — no full text of the source

Social media exacerbates distorted perceptions by disproportionately exposing people to extremists.

“There is reason to believe that social media exacerbates these issues by disproportionately exposing people to extremists.”

Affective polarization is on the rise.

“In the political domain, there is ample evidence that affective polarization is on the rise (Finkel et al. 2020, Iyengar & Westwood 2015), leading to concerns about reduced support for democratic norms (Broockman et al. 2022, Kingzette et al. 2021).”

• cites Finkel et al. 2020 sourceNot checked — no full text of the source
• cites Iyengar & Westwood 2015 sourceNot checked — no full text of the source

Motivating bystanders to denounce hostility or harassment on social media is a promising strategy.

“One promising strategy is motivating bystanders to denounce hostility or harassment on social media (Munger 2017).”

• cites Munger 2017 sourceNot checked — no full text of the source

Most people are passive on social media and only consume content.

“While most people are passive on social media and only consume content (Settle 2018), encouraging users to participate in online discourse and to react negatively to hostility may improve the social media environment.”

• cites Settle 2018 sourceNot checked — no full text of the source

Encouraging users to participate in online discourse and to react negatively to hostility may improve the social media environment.

“While most people are passive on social media and only consume content (Settle 2018), encouraging users to participate in online discourse and to react negatively to hostility may improve the social media environment.”

Our ability to generalize across platforms is severely limited.

“As such, our ability to generalize across platforms is severely limited.”

Increasing people's motivation to be accurate, and reducing people's motivation to view content through a partisan lens, improves the accuracy of people's beliefs and sharing decisions.

“Other experimental work has found that increasing people's motivation to be accurate, and reducing people's motivation to view content through a partisan lens, improves the accuracy of people's beliefs and sharing decisions (Rathje et al. 2022a).”

• cites Rathje et al. 2022aNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

These strategies often mitigate, rather than address, the root of the problem.

“However, these strategies often mitigate, rather than address, the root of the problem.”

Tweets containing moral-emotional language are more likely to receive hate speech in the replies.

“Moreover, the amount of moral convergence in online groups (e.g., people in a group sharing the same moral views) predicts radicalization (Atari et al. 2022), and, as we mentioned above, tweets containing moral-emotional language are more likely to receive hate speech in the replies (Solovev & Pröllochs 2023).”

• cites Solovev & Pröllochs 2023Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Our understanding of the consequences of moral discourse on social media may not accurately reflect the experience of users from other parts of the world or understudied communities.

“Therefore, our understanding of the consequences of moral discourse on social media may not accurately reflect the experience of users from other parts of the world or understudied communities.”

Harnessing people's social identities or social norms can motivate prosocial behavior.

“Thus, harnessing people's social identities or social norms can motivate prosocial behavior (Van Bavel & Packer 2021).”

• cites Van Bavel & Packer 2021Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Aspects of social media design, incentive structures, and algorithms can all be shifted to encourage people to share more positive content at scale.

“As such, aspects of social media design, incentive structures, and algorithms can all be shifted to encourage people to share more positive content at scale.”