The Personalization of the Internet Contributes to Political Polarization and Radicalization in North American Society

Abstract: An analysis of how the digital media experience on the Internet may lead to a more politically polarized, and even radicalized, North American society. A political radicalization framework like that used for analysis for cases of Islamic radicalization is used to analyse and find ways in which ways the present personalized Internet experience contributes to political radicalization and polarization. Other psychological factors that contribute are mentioned and discussed. Possible design considerations for Internet technology designers are included: active introduction of an opposing ideology’s ideas into a person’s social media ‘feed’, promotion of active debate among users, and devaluing time spent or amount of clicks garnered on a website as a metric of website success.

keywords: Internet, echo chambers, political polarization, political radicalization, North America, the West, ideology

1. Introduction

The browsing of the Internet has increasingly become a more personalized experience. The data collection and computational capability of Internet social and content companies has increased and so has their ability to predict a user’s preferences for what they would like to read, listen to, and watch on their screens. Undoubtedly, this has made for a more ‘comfortable’ existence online for us all, but it has an insidious cost that affects social cohesion in the long-term: the increase of political polarization and radicalization. ‘Social cohesion’ is characterized by “a set of attitudes and norms that includes trust, a sense of belonging and the willingness to participate and help, as well as their behavioural manifestations.” (Chan et al, 2006). ‘Social cohesion’ as a concept is very broad, but ‘willingness to participate and help others’ and its behavioural manifestations are severely undermined when a society undergoes increased political polarization and radicalization. We can focus on this concept of radicalization as well as the broader topic of polarization. The social and media aspects of the Internet have become a politically and ideologically radicalizing force in part because of their increasing personalization. For example, a Internet social media users’ content either posted by them or shared by them is in turn used to feed a feedback loop of determining the content that they view in the future. This may help support simple-mindedness in mild cases, and complete ideological radicalization in other cases. Although this issue has seldom been examined at a framework level, we can do analysis of Islamic terrorist radicalization and its connection to the Internet, and other political radicalization frameworks in general. This sort of examination may give us clues on how best to tackle the most important determinants for any brand of political radicalization as it relates to the social Internet and Internet media — including the liberal/conservative spectrum in the West. In the past, information was mediated to a large extent by journalists or news agencies, but we are now in the ‘disintermediated media’ era (Bessi et al, 2016). Our friends are increasingly becoming the mediators between ourselves and the truth. It is in the hands of the technology creators and designers to help curb the negative effects of the personalized Internet. At the end of this paper an attempt is made to come up with a few possible solutions to the problems raised in the paper.

2. The Internet and Echo Chambers

An echo chamber is exactly what it sounds like — you say something, and you hear the exact same thing back, in a confined space. The formation of echo chambers is not exclusive to the Internet, but the literature on the Internet’s positive effect on echo chambers’ formation is fairly conclusive. The collection of data on users and its usage to provide material that people are interested in is now accepted practice. More and more websites are predicted to force people to register to allow for this personalization to continue occur: “Publishers force more people to sign-in/register for websites and apps as well as investing heavily in data to help deliver more personalised content and messaging.” (Newson, 2017).

A RAND report also lends support to the idea that the internet may close people in to only being able to have access to their own views. This is an important factor in the formation of echo chambers:

“The consensus in the literature is that the internet allows individuals to gain easier access to the material in which they are interested, which is harder to do in the physical world where we more regularly come across individuals with different opinions or access material exposing different views (Briggs and Strugnell, 2011; Shetret, 2011).” (von Behr et all, 2013).

Echo chambers are dangerous enough on their own, but after we introduce the frameworks used to understand political radicalization and polarization, we can begin to make progress on the issue.

3. Understanding Radicalization and Polarization

There are many frameworks for understanding political radicalization and how it occurs. From a 2017 paper on the subject, there are at least 10 different frameworks identified. Of those described, we will focus on two main descriptions. The first is McCauley and Moskalenko’s own description from 2011. They identified radicalization factors at three levels: individual, group, and mass. Since we are talking about radicalization on the Internet, we will focus on the individual level. Here are the individual factors they identify: anger and revenge for harm to self or loved ones (personal grievance); outrage for injustice to a larger group or cause the individual cares about (group grievance); participation in progressively more radical acts (slippery slope); helping a friend or loved one (love); status-seeking; escapism; and “unfreezing”, which is a loss of social connection that allows the individual to be open to new worldviews and new people. (McCauley & Moskalenko, 2017). Another framework is described in as a four-stage model of radicalization: first is a “cognitive opening” to new people and new ideas following feelings of disconnection (again called “unfreezing”); second is connection with activists, either with pre-existing kin or new connections; third is the acceptance of an ideology, and finally is the justification of violence in the name of that ideology. (Wiktorowicz, 2005). Note, that the Wiktorowicz framework was originally for Islamic radicalization, but the abstract concepts are themselves described by McCauley and Moskalenko so the reference to Islam has been removed here. The factors to radicalization that can be affected by the Internet are plenty, but of special note is the access to other activists, and the importance of the emotional elements involved: ‘outrage at injustice’, ‘anger’, ‘revenge’, etc. Radicalization is an extreme concept, but another closely related concept, political polarization, has been studied recently with respect to the Internet and how it may be having an effect.

Political polarization is a broad idea about the splitting of opinion across ideological lines in larger groups. In the West, there are two main ideological groups: those who subscribe to the liberal ideology and those who subscribe to the conservative ideology. For example, it is important to note that in the past two decades, those who are consistently aligned with one or the other of these ideologies in America have more than doubled:

“The overall share of Americans who express consistently conservative or consistently liberal opinions has doubled over the past two decades from 10% to 21%. And ideological thinking is now much more closely aligned with partisanship than in the past. As a result, ideological overlap between the two parties has diminished: Today, 92% of Republicans are to the right of the median Democrat, and 94% of Democrats are to the left of the median Republican.” (Pew Research Center, June, 2014, “Political Polarization in the American Public”)

Of course, this is not to say that there is a direct causal link between the developments in Internet technology and this increased polarization, but at the very least, something should be done about this increasing polarization in America and in the West in general. With these two concepts described in more detail, we can explore how the personalized Internet and Internet in general may be contributing to their increase in Western society.

4. The Internet’s Effect on Radicalization and Polarization

Note that radicalization here refers to any set of ideological beliefs (for our purposes) — a person may become interested in a variety of topics. For now, most of the work done has been on terrorism, but new threats may emerge as the population becomes polarized, so a more abstracted understanding is necessary of the Internet’s effect on the described phenomena.

In a study of the formation of echo chambers via Twitter, their results prompted them to suggest that there is an “underappreciated ideological asymmetry” that exists in “the structure and function of online political communication.” (Barbera et al, 2015). In the same study, they tied together this concept of the formation of echo chambers and the escalation of polarization and extremism (another time for radicalism):

“[…] to the extent that individuals expose themselves to information that simply reinforces their existing views (e.g., Ditto & Lopez, 1992; Garrett, 2009a, 2009b; Sears & Freedman, 1967), greater access to information may foster selective exposure to ideologically congenial content, resulting in an “echo chamber” environment that could facilitate social extremism and political polarization (Adamic & Glance, 2005; Iyengar & Hahn, 2009; Prior, 2007)” (Barbera et al, 2015)

The first stage of radicalization mentioned in both frameworks mentioned is termed loosely as “unfreezing”, which is the formation of the cognitive opening for new beliefs because of, for example, the “perception of injustice or oppression” (commonly against non-Whites, Muslims, conservatives, liberals, and so on.) This ‘perceived injustice’ is already a common theme in Internet news articles because as human beings we value compassion. Anything that seems uncompassionate about the world is interesting enough for us to click on, especially if we are part of the group being persecuted or oppressed. If we couple this with the increased personalization of the Internet, these kinds of appeals to emotion are multiplied and feed into our radicalization and polarization frameworks. It is possible to carry out a small experiment on ones’ own computer: type into the YouTube search bar a political truth-claim about the world such as: “pro-life arguments”. For me, the recommended video list had a video called “Pro-Life Truth — 7 Ways to Silence Your Liberal Professor.” Then, in that video’s recommended videos list there were more “pro-life” supporting videos, and so on. The carrying out of empirical studies would be required to explore such ‘rabbit holes’ further, but the basic assumption seems correct: political ‘rabbit holes’ form on the Internet content websites.

If we conceptualize our Internet news and media experience as the amount of simply-stated truth-claims we perceive, then based on the example above of searching something political, we can expect there to be repetition of truth-claims at a greater rate than a person would normally encounter. The interesting thing about human psychology is that repetition is a possible way to influence somebody to believe something. The effect is called the “illusory truth effect”:

“The illusory-truth effect was first observed by Hasher, Goldstein, and Toppino (1977), who found that subjects rated repeated statements as more probably true than new statements. Repetition is an illogical basis for truth; Wittgenstein likened the tendency to believe repeated information to buying a second newspaper to see if the first one was right (Kenny, 1973).” (Begg, 1992)

Solely based on the definition of an echo chamber, this illusory-truth effect is multiplied while inside one of them, and the continued personalization of the Internet can only increase this effect and increase polarization.

An important radicalizing force is the presence of peer-group support in Internet echo chambers. As stated previously, this is a key component of political radicalization as per the framework in Wiktorowicz’ and many others’ accounts. On the Internet, especially such a personalized version, you can find a multitude of different groups online to garner support from. It is possible to use this as a source of social good. It is also possible to embed yourself into a support system of opinions that are unsupported by fact or that are destructive. If a user of social media watches certain videos that repeat certain truth-claims to him, he will be served more videos with those truth claims in them as well. Through the process of personalizing the web experience, it also makes self-radicalizing easier if a person’s interests start to get political. Their web experience teaches them rhetoric, argumentation styles, and perceived injustices that must be remedied. The experience becomes one where “brought together by online journals, blogs, services and chatrooms, the participants enter forums where the extremist ideology becomes self-reinforcing.” (von Behr et al, 2017).

An echo chamber whereby the values and worldviews of the agents are so similar as to not be able to have discussions that disagree with the narratives agreed upon. It is a dynamic process that breeds radicalised viewpoints and strengthens potentially false truth-claims about the world in the minds of the individuals inside of them. Little opposing evidence can get through into the chamber. Usage of websites like Facebook involves actions such as ‘liking’ or ‘following’ other users’ content or ‘sharing’ content they find to be compelling or agreeable. People tend to share and like content with which they agree on the ideological level: “When it comes to explicitly political issues, individuals are clearly more likely to pass on information that they have received from ideologically similar sources than to pass on information that they have received from dissimilar sources. (Barbera, 2015).” Moreover, algorithms in websites such as Facebook and YouTube keep track of these sorts of actions, and change the type of content that is displayed: “[…] on online social media, different algorithms foster personalized contents according to user tastes — i.e. they show users viewpoints that they already agree with.” (Bessi et al, 2016).

This inclination to expose yourself to content you agree with ideologically is called “selective exposure”, and this inclination is more powerful than the wilful avoidance of ideologically dissimilar viewpoints. The motivation of websites such as Facebook for this is to keep users engaged with the service as much as possible to increase the value for advertisers — views of their ads and clicks on their ads. However, this motivation and its effects on consumer’s psychology and ideological stances had not been examined before its instantiation and we are now seeing the results in society. The World Economic Forum listed ‘massive digital misinformation as one of the threats to modern society.’ On a heavily ideologically personalized Internet, misinformation can much more easily permeate through a network inside echo chambers.

The Internet and social media also provide an element of anonymity and protection from detection, which may help people pursue social connections to more extreme activists and to commit unethical acts.

“The internet provides supposed anonymity (Weimann, 2006) and a degree of protection and security from detection (Gray and Head, 2009). It also provides acceptance: information is non-censured and non-hierarchical” (Von Behr et al, 2017).

In the ‘real world’, it is much harder for a person to take action, but the Internet gives them the opportunity to participate and, in the very least, encourage others to believe as they do. For women in Islam, this may manifest as the ability to participate in discussions with other men online under the anonymity provided by the Internet. For social movements in general, it may mean that a social anxious individual may contribute to a movement rather than be shunned away or be too anxious to get involved in real life. (von Behr et al, 2017). Although these are separate issues from the personalization of the Internet experience, it is an added burden and issue to solve for reducing the potential of the Internet for polarization and radicalization.

The power of online and social media to affect views is also more powerful than that of traditional media because we are more likely to believe that something is true if one of our friends tells us something than if we hear it on the television or in the newspaper. The continued personalization of the Internet also makes this into a larger problem for its future as a beneficial social tool. These and other issues described may be difficult to solve, but some implications may be concluded for the future Internet technology designer and the industry at large.

5. Implications for the Design of Technology for the Internet

Already, there have been calls for technology to combat the problems with the Internet described in this paper. From a 2017 report on technology, so-called “Serendipity Algorithms” are predicted:

“Serendipity algorithms: Expect to see the creation and marketing of services that challenge our prejudices; that give you more of what you don’t like. The Guardian has already started a feature: ‘Burst your bubble’, which offers links to conservative articles that thoughtful liberals should read. But countering the bubble, says City University’s Jane Singer, will be harder than it seems given “how viscerally it feeds our emotions and perceptions, and the extent to which every one of us exists within it”. (Newson, 2017).

This quote touches directly on the central problem: the social Internet has become an ideological manipulation machine that every one of us exists within, and more than that, it is in the incentive structure of both the suppliers (the technology companies) and customers (individuals browsing the Internet) to continue developing more personalized search results, video recommendations, shared posts from our friends, and so on.

For designers of technology, it is important to take these ideas into consideration. The social Internet has become a contributing force in the political division in Western society and elsewhere. This is combined with the added possibility that non-Western societies may actively be attempting to upset the political structures of Western society by attempting (and succeeding) in ideologically subverting the West into believing certain truth-claims and disbelieving other ones, or just generally creating an atmosphere of uncertainty and chaos. (Vicario et al, 2016). Ideological subversion is the term used to describe this process: the manipulation of public opinion on issues that affect social cohesion.

Much work is to be done on this issue, but I may propose one possible solution for a content website such as YouTube. To guard against the ‘falling down a rabbit hole’ effect when it comes to central issues such as religion and political opinion, a real-time analysis of every YouTube user may inform an algorithm to either promote or demote certain videos in the ‘recommended’ videos tab. Similarly, if Facebook were to occasionally promote thoughtful ideologically dissimilar content regardless of user preference, perhaps with the headline “Have your say on this video!”, then perhaps people would encounter well-thought-out opinions and other ideas than they may not have had the chance to come across in their personalized Internet browsing.

Using radicalization and polarization frameworks, it is possible to make progress on deciding what issues we need to tackle going forward with Internet technology. The Internet can be used as a tool for reducing these phenomena, but the incentive to personalize a user’s experience to keep them browsing certain websites is very strong. It is only through increasing the value of social cohesion over that of ‘views, ‘clicks’, and ‘time spent on page’ that this issue can be kept at bay, and to reduce the potential of polarizing and radicalizing the West to a point of no return.

References:

Barbera, P., Jost, J. T., Nagler, J., Tucker, J. A., & Bonneau, R. (2015). Tweeting From Left to Right. Psychological Science,26(10), 1531–1542.doi:10.1177/0956797615594620

Begg, I. M., Anas, A., & Farinacci, S. (1992). Dissociation of Processes in Belief: Source Recollection, Statement Familiarity, and the Illusion of Truth. Journal of Experimental Psychology,121(4), 446–458. http://dx.doi.org.proxy.lib.sfu.ca/10.1037/0096-3445.121.4.446

Bessi, A., Zollo, F., Del Vicario, M., Puliga, M., Scala, A., Caldarelli, G., . Quattrociocchi, W. (2016). Users Polarization on Facebook and Youtube. PloS One, 11(8), E0159641.

Chan, J., To, H., & Chan, E. (2006). Reconsidering Social Cohesion: Developing a Definition and Analytical Framework for Empirical Research. Social Indicators Research,75(2), 272–302. doi:https://doi-org.proxy.lib.sfu.ca/10.1007/s11205-005-2118-1

Hawkins, S. A., Hoch, S. J., & Myers-Levy, J. (2001). Low-Involvement Learning: Repetition and Coherence in Familiarity and Belief. Journal of Consumer Psychology,11(1), 1–11. doi:10.1207/S15327663JCP1101_1

McCauley, C., Moskalenko, S., Kazak, Anne E., & Horgan, John G. (2017). Understanding Political Radicalization: The Two-Pyramids Model. American Psychologist, 72(3), 205–216.

Michela Del Vicario, Alessandro Bessi, Fabiana Zollo, Fabio Petroni, Antonio Scala, Guido Caldarelli, Walter Quattrociocchi. (2016). The spreading of misinformation online. Proceedings of the National Academy of Sciences, 113(3), 554–9.

Newson, N. (2017). JOURNALISM, MEDIA, AND TECHNOLOGY TRENDS AND PREDICTIONS 2017 (pp. 1–35). Oxford, United Kingdom: Reuters Institute for the Study of Journalism.

von Behr, I., Reding, A., Edwards, C., & Gribbon, L. (2013). Radicalisation in the digital era. RAND Corporation. Retrieved October 09, 2017.

“Political Polarization in the American Public: How Increasing Ideological Uniformity and Partisan Antipathy Affect Politics, Compromise and Everyday Life” Pew Research Center, Washington, D.C. (June 12, 2014). http://www.people-press.org/2014/06/12/political-polarization-in-the-american-public/