Thread by @boralexander1

🚨New publication alert🚨 Our paper with @M_B_Petersen has been accepted at American Political Science Review @apsrjournal. "The Psychology of Online Political Hostility: A Comprehensive, Cross-National Test of the Mismatch Hypothesis". psyarxiv.com/hwb83/ Thread 🧵👇

@M_B_Petersen @apsrjournal There appears to be a public consensus that online discussions about politics are terribly hostile, more so than offline discussions. We show evidence of such ⭐️hostility gaps⭐️ from two representative samples in USA and Denmark. But why the large asymmetry? /2

Prominent explanations emphasise a #mismatch between the evolved mind and the novel features of online communication environments: no non-verbal cues, no personal reputations (large networks, anonymous accounts, strangers hopping from one community to another). /3

Perhaps people are hateful online by accident struggling to contain their emotions in this novel environment? We outline 3 hypotheses: Hyp A) people ⭐️change⭐️ their behavior when discussing politics online -- even otherwise nice people turn hostile. /4

Surprisingly, we found no evidence for this hypothesis. Across four representative samples, we find remarkably high correlations between self-reports of online and offline political hostility. The people hateful on Twitter offend others in face-to-face conversations too. /5

Hyp B) Hostile people ⭐️select into⭐️ online discussions. So it's not nice people behaving badly, but bad people occupying online platforms. We find mixed evidence: Hostile people talk a lot about politics both on- and offline, and nice people select out of online env. /6

Caveat: the latter result (nice people's selecting out) did not replicate in our latest US sample, which we collected already during covid-19. We can't be sure if it's the pandemic or something else. (more research is needed!) /7

Hyp C) People ⭐️perceive⭐️ hostility to be worse online. While so far we focused on offenders, perhaps the problem is that victims tend to suffer more from protracting, public online quarrels. But our data shows that if anything offline debates are more taxing emotionally. /8

We also formulate an alternative to the mismatch explanations: the ⭐️connectivity⭐️ hypothesis. This too rooted in #evopsych: dominating others by intimidation and coercion is a sure way to gain status. /9 (see @Joey_Cheng_ et al)

@Joey_Cheng_ Perhaps status-seeking individuals are offensive in all environments and the bad PR of online environments is due to mechanical effects of increased connectivity and visibility. /10

@Joey_Cheng_ Put differently, while most worry about online echo chambers, paradoxically the problem may be too much exposure to people, we would reliably avoid in the offline world (echoing recent arguments by @kmmunger and @chris_bail). /11

@Joey_Cheng_ @kmmunger @chris_bail Our results are most consistent with this explanation. Status-driven risk taking is a very reliable predictor of both online and offline hostility and frequency of talking about pol. /12

Besides, people report most asymmetry when witnessing hostility against strangers (as opposed to self and friends) /13

To sum up, we find no evidence that mismatch is a major factor behind online hostility. Check out the full paper for more theory and results (we report 8 studies with total N = 7,510) and some caveats. psyarxiv.com/hwb83/ /14