Houjun Liu

PWR1 Texts in Conversation Planning

Quotes Bin

polarization distorts beliefs about others

“Recent years have seen a sharp increase in political polarization in the United States (1–7), leading to deadlock in Congress (8), distorted beliefs about fellow Americans (9, 10), and distrust, hostility, and even violence toward outgroup members (11–13)” [Novoa et al., 2023, p. 1]

generics about particular group is a way that polarized languages manifest

“Specifically, we focus on expressions that make claims about a category as a whole (e.g., “Democrats want to defund the police” makes a claim about the category of “Democrats”), also known as generics (34–38).” [Novoa et al., 2023, p. 2]

generics are much more readily remembered

“A second distinctive signature of generics documented in prior research is that they tend to be how generalizations are later recalled, even when generalizations are stated in more precise, quantified ways” [Novoa et al., 2023, p. 2]

generics are strongly rejected or accepted based on parity affiliation => echo chamber

“Respondents showed a strong pattern of accepting generics for the target party and rejecting generics for the opposite party.” [Novoa et al., 2023, p. 3]

people perceive higher polarization than actually are present due to use of generics

“We found that perceived polarization was greater than actual polarization, in two key respects: 1) for nearly every issue, people believed that the two parties were further apart than they actually are, and 2) patterns of generic endorsement were more polarized (i.e., revealed a greater gap between the two parties) than perceptions of prevalence.” [Novoa et al., 2023, p. 5]

polarized language is present far more in generic statements

“The results of this study support the conclusion that generic language leads to polarized judgments regarding political parties, and does so more than nongeneric language. We obtained three key results: 1) for generic statements (e.g., “Democrats …”), prevalence estimates were larger for the named party (e.g., Democrats, when the generic statement was about Democrats) than for the unnamed party (e.g., Republicans, when the generic statement was about Democrats); 2) for generic statements, prevalence estimates were above 50% for the named party and below 50% for the unnamed party; and 3) the gap between named and unnamed prevalence estimates was larger for generic statements than for nongeneric statements (such as “Many Democrats support House Bill 858” or “Some Democrats support House Bill 858”).” [Novoa et al., 2023, p. 9]

political science generally believes that political polarization and citizen polarization is different

“The predominant view in political science is that the current polarization in Congress has not diffused much into the citizenry” [“Political Polarization and the Dynamics of Political Language: Evidence from 130 Years of Partisan Speech [with Comments and Discussion]”, 2024, p. 5]

political polarization and its prevalence has been discussed since founding of the country

“The idea that U.S. politics is necessarily polarized, owing to the intrinsic diversity and size of the country, goes back at least to James Madison and the divergence between Hamiltonian and Jeffersonian economic philosophies” [“Political Polarization and the Dynamics of Political Language: Evidence from 130 Years of Partisan Speech [with Comments and Discussion]”, 2024, p. 5]

partisanship switches to minority party when the house switches control

“the partisanship of language tends to switch when House control switches, but in the direction of the new minority party.” [“Political Polarization and the Dynamics of Political Language: Evidence from 130 Years of Partisan Speech [with Comments and Discussion]”, 2024, p. 25]

one type of polarization is where people disengage with those with opposite views

“The first aspect of political polarization, which we call “interactional polarization,” focuses on a process whereby participants in a debate increasingly interact with likeminded individuals, while disengaging from interactions with others who hold opposing viewpoints.” [Yarchi et al., 2021, p. 101]

there’s a difference between filter bubbles and increased polarization

“Despite the recent salience of theories regarding fragmented “echo chambers” or “filter bubbles,” it remains contentious whether social media do indeed drive such interactional polarization” [Yarchi et al., 2021, p. 101]

already known groups typically bring strong agreements, and strangers typically bring disperate views

“Heterophilic interactions appear to be more common along so-called “weak ties” occasional communications that are not underfed by strong social bonds such as friendship or sustained collaboration – while most “strong ties” (among friends, within teams, etc.) are predominantly homophilic” [Yarchi et al., 2021, p. 101]

social media users agree more over time

“H2 (Interactional Polarization): Interaction patterns on social media become increasingly homophilic over time.” [Yarchi et al., 2021, p. 102]

isolation results in more extreme contributions

“Individuals embedded within more homophilic interaction networks subsequently express more extreme positions in their contributions.” [Yarchi et al., 2021, p. 102]

Facebook’s chief innovation is to leverage the in-group homogeneity to create supportive opinions

“As the world’s foremost social media platform, Facebook’s popularity is arguably derived largely from its capacity to immerse its users in a feed of contents that cater to their personal interests and leanings. To do this, the platform relies heavily on users’ self curated networks of friends, but also on an algorithm that prioritizes content based on users’ interests and support for similar posts, displaying only a small share of predominantly congenial, supportive contents” [Yarchi et al., 2021, p. 104]

Facebook creates supportive echo chambers

“Facebook has become the prime suspect for the creation of homophilic echo chambers” [Yarchi et al., 2021, p. 104]

Twitter creates asymmetric, non-friend dynamics

“Twitter is defined primarily by its unrestricted publicness. Anyone, even non-users, can read any tweet, and any user can respond to any contribution. Users can follow others without a need for permission, enabling asymmetric, non-reciprocated ties.” [Yarchi et al., 2021, p. 104]

Debate exists between whether Twitter creates or dismantles homogenization, and therefore polarization

“Reflecting Twitter’s ambiguous profile, the existing literature yields conflicting findings regarding its tendency toward homophily and polarization (e.g., Kwak et al., 2010 detected little homophily; Weng et al., 2010; Hong & Kim, 2016; Colleoni et al., 2014 found the opposite).” [Yarchi et al., 2021, p. 105]

CITE: defines social media

“social media platform in a narrow sense (following Ellison and Boyd (2013) definition)” [Yarchi et al., 2021, p. 105]

Users tend to express more extreme views if surrounded by likeminded users

“Considering the effect of homophilic interactions on expressed positions, our data confirm users’ tendency to express more extreme views if interactions with likeminded users take in a larger share of their social media communications (H3).” [Yarchi et al., 2021, p. 111]

cannot study only one social media as they have different properties

“Beyond questioning the widespread reliance on Twitter (and limited public Facebook) data to draw conclusions about social media as a whole, our study also highlights the perils of inferring dynamic properties from static data.” [Yarchi et al., 2021, p. 114]

tendency to associate with like-minded people increases echo chambers

“That social psychology has long shown this tendency to associate with like-minded others is common cross-culturally. However, there is new fear that the current media system is helping people enter echo chambers more easily than ever before.” [Dubois and Blank, 2018, p. 731]

Twitter is an isolated slice of the population

“Twitter itself is used by a relatively small proportion of the population, about one-quarter of the UK, which is younger, wealthier, and better-educated than Britain as a whole” [Dubois and Blank, 2018, p. 732]

in the UK, going to a news source like BBC is still more common

“going directly to a news source such as the BBC remains more common in the UK (Newman et al., 2017).” [Dubois and Blank, 2018, p. 733]

studies don’t study the aggregate effect of diverse media

“A core problem with this line of research is that most studies select only one or a few media to focus on and so the comparative utility or effects of use of media in a diverse media environment are unclear.” [Dubois and Blank, 2018, p. 733]

people with strong partisanship report consuming a diverse media digest

“First, even individuals who have strong partisan affiliation report using both general news sites (which are largely non-partisan and include a variety of issues) and niche news sites (which may be partisan or focused on specific issues) – Republicans and Democrats have media diets which are quite similar” [Dubois and Blank, 2018, p. 734]

consumption of mixed media results in incidental exposure to a variety of news sources

“While one might receive primarily left-leaning political content on Twitter, they may be incidentally exposed to a right-leaning perspective from a family member on Facebook or they might hear a debate between representatives from various perspectives on a television news broadcast.” [Dubois and Blank, 2018, p. 734]

those who are politically aware are going to encounter more perspectives

“As Prior argues, political ‘junkies’ are likely to consume a lot of information and therefore may encounter more perspectives and arguments” [Dubois and Blank, 2018, p. 734]

increased involvement in politics results actually in less echo chamber

“H2: The higher a person’s level of political interest the less likely they are to be in an echo chamber” [Dubois and Blank, 2018, p. 735]

People who are actually disinterested in politics are in an echo chamber

“First, that respondents with no political interest are in an echo chamber. We examine this possibility using the regressions in Table 3. The results in this table are based only on the respondents who said they had ‘No interest at all’ in politics, N = 243.” [Dubois and Blank, 2018, p. 739]

High choice in media doesn’t mean a high degree of ability to reconsiliate

“A high-choice media environment does not simply mean that individuals develop strategies to deal with the many media options available, though of course they do so as they develop their news and political information repertoires” [Dubois and Blank, 2018, p. 740]

diversity in media AND engagement in politics matters

“Our results suggest that people who are both not politically interested and who do not use diverse media are more likely to be in an echo chamber. They are less likely check multiple sources or to discover things that change their minds.” [Dubois and Blank, 2018, p. 741]

polarized language results in greater engagement: but only to trolls and politically engaged users

“We also find that polarized language is associated with greater engagement, but this association only holds for politically engaged users (both trolls and regular users). This research clarifies how trolls leverage polarized language and provides an open-source, simple tool for exploration of polarized communications on social media.” [Simchon et al., 2022, p. 1]

Russian trolls used more polarized language

“Again, we find that politically oriented Russian trolls use significantly more polarized language than their politically matched American sample (Russian trolls: M = 5.16, SD = 8.00, and N = 55,726; American controls: M = 2.91, SD = 6.84, and N = 55,726), t(108,836) = 50.61, P < 0.001, and Cohen’s d = 0.30 (for a robustness check, see Supplementary Materials).” [Simchon et al., 2022, p. 4]

foreign agents increase in their polarization and posting frequency

“foreign agents from various countries strategically used polarized language in social media communications, and in a majority of cases we see an increase over time in these attempts.” [Simchon et al., 2022, p. 6]

distinction between polarization on issues vs. polarization of anger

“Scholars have made the conceptual distinction between issue polarization—an ideological, policy-based political divide, and affective polarization, i.e. dislike, distrust, and general animosity of political partisans toward the other political side” [Simchon et al., 2022, p. 6]

small amount of trolls can polarize lots of people

“Questions remain as to the extent of influence of trolls’ social media presence on real people. However, it is important to note that even a small number of agents with aggressive attitudes can have a substantial influence on the majority view, a process called “information gerrymandering”” [Simchon et al., 2022, p. 9]

interaction with trolls didn’t seem to change partisanship

“The authors found that only a small fraction of users interacted with Russian trolls, and they did not observe any change in partisan attitude during that time among these users (68).” [Simchon et al., 2022, p. 9]

subclaim organization

Polarization comes from congenial echo-chambers driven by generic language, which social media is prone to create due to their curation

Interaction with in-group only results in more extreme contributions. This is what we typically call an “echo chamber” (1, 2). One such chamber enviroment is social media, a particularly salient case of this is Facebook, which creates supportive echo chambers.

Novoa proposes one analysis through linguistics by which such an echo chamber can get created—generics: easy to remember generalisations. polarized language is present far more in generic statements, and generics about particular group is a way that polarized languages manifest.

Generics only function when deployed within a homegenous environment. Yet, others have noted that Facebook’s chief innovation is to leverage the in-group homogeneity to create supportive opinions—displaying “congenial” content that are likely to be homegenous.

The congenial environment itself, however, is not enough to create or disrupt polarization; breaking echo chambers requires both a diversity of opinions as well as actual engagement

Unlike Facebook’s congeniality, Twitter creates asymmetric, non-friend dynamics. Though it shows that it helps dismantle some echo chambers, its not conclusive. Yarachi notes this as the difference between filter bubbles and increased polarization. A “filter bubble” itself isn’t polarization, so what is?

Dubois solves this mystery by arguing that it is people who are actually disinterested in politics are in an echo chamber. Through consumption of mixed media results in incidental exposure to a variety of news sources one has to participate in the conversation to get out the echo chamber.

Meaning, high choice in media itself (i.e. having facebook AND twitter) doesn’t mean a high degree of ability to reconsiliate. It is diversity in media AND engagement in politics matters.

other notes

By using polarized language to target only politically active users, trolls essentially disrupt the ability to dismantle echo chambers

Interaction with trolls didn’t seem to change partisanship, yet previous work establishes that a small amount of trolls can polarize lots of people—so the manner by which trolls work is confusing.

Simchon notes that Russian trolls used more polarized language. Our previous analysis concludes that political activism is an important and inseperable part of breaking an echo chamber; trolls, then take advantage of this fact to disrupt the process of breaking away from polarization by capturing already politically active users, which trolls take part.

BIN