Filter Bubbles Research Articles

"Virtual Parallel Societies"—A Study on the Differences in Mobile Application Usage Between Hong Kong and Mainland Chinese Adolescents

Objective: This study explores the differences in mobile application usage between adolescents in Hong Kong and Mainland China, highlighting how these variances contribute to the formation of virtual parallel societies. Methods: The theoretical framework incorporates Walter Lippmann's Public Opinion and Eli Pariser's Filter Bubbles to explain how algorithm-driven platforms foster informational silos. The study employed a mixed-methods approach, primarily using secondary data from the QuestMobile 2023 China Mobile Internet Annual Report and data from the Department of the Government of the Hong Kong Special Administrative Region, along with qualitative research methods such as semi-structured interviews and content analysis of social media interactions. Results: The findings suggest significant disparities in app usage between adolescents in Hong Kong and Mainland China, influenced by technological, sociological, and political factors. Adolescents in Hong Kong preferred global platforms like Instagram and Facebook, while Mainland Chinese adolescents used local platforms like WeChat and Weibo. These differences were influenced by sociological factors, including class, education, and access to technology, and contributed to the formation of distinct informational silos. Conclusion: The differences in mobile application usage have led to the development of virtual parallel societies, with implications for national identity and socio-political cohesion. Policy recommendations include enhancing media literacy, promoting Chinese perspectives globally, and encouraging social media platforms that foster a shared national identity.

Missed encounters: what may be relevant for an AI is not for a human being

Abstract The World Wide Web has been a fundamental part of our daily lives for years. Its algorithmic framework ensnares our online journeys in an “endless recurrence” of the “same” by creating multiple filter bubbles. Digital algorithms establish a precise “order of discourse,” leaving little to no room for deviation. Functioning as a colossal machinic apparatus, the web embodies the culmination of Artificial Intelligence (AI), transforming every piece of posted content into a database that profiles our online behavior and activities. This article explores whether approaches that describe everyday human communication, such as the theories of relevance developed by Alfred Schutz, or by Dan Sperber and Deirdre Wilson, can be applied to the realm of digital algorithmic grammars governing the web, and to the semiospheres that animate it. The conclusions raise some doubts: a gap exists between algorithmic language and its historical counterparts, characterized by a disparity between logical-mathematical grammar and other linguistic-historical-natural ones. While these two types of languages coexist within the digital landscape, their relevance differs; what may be “relevant” in an algebraic context may not necessarily translate to our live conversational exchanges, and vice versa. Although merged in the digital sphere, these languages operate distinctly, and proficiency in one does not guarantee adeptness in interpreting the other accurately.

Testing the effects of personalized recommendation service, filter bubble and big data attitude on continued use of TikTok

PurposeThis study attempts to identify the determinant factors for Chinese users’ intention of continuous use of TikTok.Design/methodology/approachDrawing on the theoretical background and previous literature, this study proposes the research model followed by research hypotheses. For the purpose of empirical analysis of this study, we conducted a questionnaire survey on current Chinese TikTok platform users. Factor analysis was conducted to verify the reliability and validity of the measurement variables; Confirmatory factor analysis was conducted to ensure that the structural dimensions of the measurement are effective; the analysis results of the structural equation model for hypothesis test.FindingsThe result of the study indicates that the core service traits (entertainment, informativeness and convenience of use) of TikTok all have a positive effect on user engagement and perceived usefulness. In addition, the study found that both personalized recommendation services and filter bubble, which are features of TikTok's service recommendation system, have a positive effect on user engagement and perceived usefulness of TikTok. In addition, it was found that both user engagement and perceived usefulness have a significant positive effect on continuous use intention. Finally, users’ attitude toward big data indicating privacy concern only had a positive moderating effect on the personalized recommendation service but not on filter bubble.Originality/valueAs consumers’ demand for services tailored to their needs and preferences fast increased, it calls for growing concern over the safety of personal data being used by the platforms. Although this paradox over increased user convenience versus privacy protection issue has been persistently recognized, no previous research empirically addressed this issue in the context of social media platforms.

Issue agenda-setting in the voice to parliament referendum: using big data to explain voice discourse on traditional and social media

ABSTRACT This article explores issue agenda-setting in Australia's 2023 Voice to Parliament referendum, analysing the dynamics of Yes and No campaigns in a diverse media ecosystem. We identify influential messages across this complex media environment using soft clustering of sentence embeddings from large language models (LLMs). Drawing on 3.3 million documents and over 10 million Voice specific sentences from mainstream and social media sources, we identify the narrative configurations of the campaigns, their audience size and engagement. Overall, the Yes campaign dominated with more diverse messaging and greater audience size and engagement than No. Yet, the No case prevailed. We find the narrower, mostly negative No campaign helped sustain the issue agenda both online and offline, forcing Yes to respond to it. The findings also challenge echo chamber theory, revealing most unique topics contained content from both #Yes and #No perspectives, indicating robust public engagement rather than isolated filter bubbles.

Hydrodynamics of molten media bubble columns for hydrogen production through methane pyrolysis

Methane pyrolysis using a molten media bubble column reactor is a promising technique for hydrogen production with low carbon dioxide emissions at a feasible price. Understanding the bubble dynamics in molten media is essential to elucidate the reaction mechanisms and establish design requirements for efficient reactors. Computational fluid dynamics provides an effective means to understand the hydrodynamics in opaque molten media. This research used the volume of fluid method to study the effects of gas injection rate as well as variations in gas and molten media (iron, aluminum, and a salt mixture of sodium bromide and potassium bromide in a 48.7:51.3 molar ratio) properties on bubble dynamics. The computational model was first validated using existing experimental and empirical observations. This study makes fundamental contributions to the understanding of bubble dynamics in molten media. First, it was confirmed that gas properties had a small effect on bubble dynamics. The difference in bubble diameters between argon at ambient temperature and 1600 °C was less than 10%. Second, it was found that the volumetric gas injection rate and molten media properties significantly impacted the bubble dynamics, including the bubble diameter and flow regime. Future work will build on these findings to recommend appropriate operating conditions and molten media for specific pyrolysis reactor designs.

Weaponization of Social Media – Challenges and Responses

The use of social media as a tool for manipulation poses significant challenges worldwide, affecting public sentiment, intensifying divisions, and endangering democratic systems. This article investigates these issues and examines the responses from governments, tech companies, and civil society. The swift evolution of social media has changed the way we communicate while also facilitating the rapid dissemination of false information and polarizing narratives, threatening societal unity. Researchers emphasize the impact of algorithms in creating echo chambers and filter bubbles that deepen polarization. Additionally, malicious actors take advantage of these platforms to propagate disinformation, sway elections, and incite violence. In response, initiatives such as enhanced fact-checking, media literacy programs, and better content moderation by social networks are being implemented. Regulatory actions to promote transparency and accountability are also recommended, alongside international collaboration to combat cross-border online threats. A comprehensive strategy involving technology, regulation, international co-operation and community resilience is essential to protect democratic ideals in our interconnected society.

The Information Isolation Trope: Isolation, Infection, and Information Silos in Early American Literature

Many early American novels feature at least one character who becomes physically isolated, is restricted access to communication, becomes ill, spreads their sickness, and then either dies or becomes permanently disabled. This repeated pattern of isolation, infection, information, and death/disability underscores just how concerned early Americans were with the potentially harmful effects of information silos. The concern with information siloing, at first glance, seems anachronistic to early America. Eli Pariser popularized the concept of a “filter bubble” in 2011 when lamenting the many ways that algorithms create echo chambers and political fragmentation. While this essay concurs that content-recommendation bubbles are exacerbated by new media technologies, early American novels suggest that people have been worrying about the regulation and isolation of information since the dawn of the new republic. Early American novels—particularly those published between 1789 and 1815—are often read as seduction narratives, didactic narratives, or both; this article argues that attending to this pattern offers a new way of reading early American fiction that underscores just how concerned readers and writers were with the potentially harmful effects of information silos and the regulation of information on the body, the body politic, and public health.

Enhancing Reflective and Conversational User Engagement in Argumentative Dialogues with Virtual Agents

In their process of information seeking, human users tend to selectively ignore information that contradicts their pre-existing beliefs or opinions. These so-called “self-imposed filter bubbles” (SFBs) pose a significant challenge for argumentative conversational agents aiming to facilitate critical, unbiased opinion formation on controversial topics. With the ultimate goal of developing a system that helps users break their self-imposed filter bubbles (SFBs), this paper aims to investigate the role of co-speech gestures, specifically examining how these gestures significantly contribute to achieving this objective. This paper extends current research by examining methods to engage users in cooperative discussions with a virtual human-like agent, encouraging a deep reflection on arguments to disrupt SFBs. Specifically, we investigate the agent’s non-verbal behavior in the form of co-speech gestures. We analyze whether co-speech gestures, depending on the conveyed information, enhance motivation, and thus conversational user engagement, thereby encouraging users to consider information that could potentially disrupt their SFBs. The findings of a laboratory study with 56 participants highlight the importance of non-verbal agent behaviors, such as co-speech gestures, in improving users’ perceptions of the interaction and the conveyed content. This effect is particularly notable when the content aims to challenge the user’s SFB. Therefore, this research offers valuable insights into enhancing user engagement in the design of multimodal interactions with future cooperative argumentative virtual agents.

The Influence of Identity Bubble Reinforcement on the Happiness Index among Chinese Medical Staff: The Mediating Role of General Self-Efficacy.

This study aimed to explore the mediating role of general self-efficacy in the influence of social media identity bubble reinforcement on the happiness index among Chinese medical staff. This study utilized data (n = 877) from the 2022 Psychological and Behavioral Investigation of Chinese Residents (PBICR) related to medical staff. Correlation analysis was performed using SPSS 26.0 software, and the relationships between general self-efficacy, social media identity bubble reinforcement, and the happiness index of Chinese medical staff were examined using AMOS 23.0 software. The happiness index of Chinese medical staff was scored at 2.93 ± 1.21 for each item. Social media identity bubble reinforcement among Chinese medical staff was positively correlated with general self-efficacy (r = 0.380, P < .001), as well as with the happiness index (r = 0.330, P < .001). General self-efficacy was positively correlated with the happiness index (r = 0.575, P < .001) and was found to mediate the relationship between social media identity bubble reinforcement and the happiness index of Chinese medical staff. In terms of the mediating effect of social media identity bubble reinforcement, the direct effect's 95% confidence interval (CI) was 0.031-0.142, accounting for 34.68% of the total effect, which is statistically significant (P = .015), while the 95% CI of the indirect effect was 0.125-0.212, representing 65.32% of the total effect, also statistically significant (P = .005). Social media identity bubble reinforcement and general self-efficacy were found to be important factors influencing the happiness index of medical staff. Hospital managers and policymakers can thus enhance the social media identity bubble reinforcement and general self-efficacy of Chinese medical staff through effective measures to improve their happiness indices.

Repetitive shock waves generated by a single long pulse underwater arc discharge

The electrohydraulic effect induced by underwater arc discharge is an efficient way to generate controllable, high-intensity shock waves. However, the development process of underwater arc discharge involves the complex coupling of plasma arc, gas bubble, and liquid medium, of which the evolution mechanism is not well understood. In this paper, the underwater arc discharge process at a millisecond pulse (&amp;gt;50 ms) was investigated by high-speed shadow imaging and colorimetric temperature measurement, and a simulation model of bubble pulsation was proposed to quantitatively estimate the state variation and energy transfer of the gas bubble. The results indicate that the whole arc discharge process can be categorized into three successive stages: short-period oscillation, long-period oscillation, and quasi-steady state. The vapor inside the bubble can reach a supercritical state (827 K and 140 MPa) at the minimum bubble radius. The simulation shows that the light radiation absorption and the heat conduct loss through metal electrodes are the two dominant factors influencing the pulsation of the bubble, and further analysis indicates that the dynamic evolution of the arc determines the bubble pulsation mode. Our findings demonstrate why and how repetitive electrohydraulic shock waves can be generated by a single long pulse underwater arc discharge, providing a low-cost way of shock wave generator based on an AC/DC high-voltage power source.

Communities Influence Social Media in Dealing with ‘Filter Bubbles’ and ‘Echo Chambers’ in The Era of Digital Content Among Students

Communities Influence Social Media in Dealing with ‘Filter Bubbles’ and ‘Echo Chambers’ in The Era of Digital Content Among Students

Students' awareness and conceptions of science‐related communication mechanisms on social media

AbstractThe rise of social media platforms and subsequent lack of traditional gatekeeping mechanisms have enabled the proliferation of scientific disinformation. Users attempting to properly evaluate scientific information and disinformation are immensely obstructed by media communication mechanisms such as filter bubbles and echo chambers. Given the recent approaches to reconceptualizing the nature of science represented by facets of media communication mechanisms, we report results from seven focus groups of 26 tenth‐grade students (M = 15 years; 58% female, 38% male, 4% nonbinary) investigating such mechanisms as applied to climate change content on social media. Using qualitative content analysis, we identified the students' awareness and conceptions of mechanisms such as filter bubbles and echo chambers on a continuum between rather simple and elaborate. The findings suggest that the students have a general awareness of most mechanisms based on their own experiences and describe additional media communication mechanisms, such as bots, microtargeting, and, particularly, algorithms, which appear specifically relevant in dealing with scientific disinformation on social media. Based on the results, we derive a set of needs for science educational learning materials and science media literacy to prepare students to tackle scientific disinformation on social media.

Data Collaborative Contrastive Recommendation model with self-adaptive noise

The recommender system recommends items to the users based on their preferences of implicit feedback. However, implicit feedback often contains noise that deviates from the user’s true preferences, thereby influencing the accuracy of the recommendations. The most effective denoising method is Self-Guided Denoising Learning (SGDL), providing a general denoising scheme that can be applied to various recommendation models. However, it is typically hard to capture user preferences efficiently and enhance the diversity of recommendations to mitigate the ‘filter bubble’ phenomenon. To address these challenges, we propose a novel Data Collaborative Contrastive Recommendation model with self-adaptive noise (DCCR). Specifically, we design the informative item extraction module to mine informative items from the original interactions and improve the accuracy and diversity of recommendations by collaborative training of the informative and the original dataset to learn diverse user embeddings and adapt to noise. Extensive experiments on three public datasets demonstrate the superiority of our DCCR over state-of-the-art methods, balancing the diversity of recommendation lists while optimizing recommendation accuracy.

Comment on the Article “Filter Bubbles and the Unfeeling: How AI for Social Media Can Foster Extremism and Polarization” (PHTE-D-23-00352R3)

Comment on the Article “Filter Bubbles and the Unfeeling: How AI for Social Media Can Foster Extremism and Polarization” (PHTE-D-23-00352R3)

Filter Bubble Trap: The Effect of TikTok Media on Language Use and Interaction Among Communication Science Students at UINSU

This article aims to explore the impact of the filter bubble created by the social media algorithm of TikTok on the language use and social interactions of Communication Studies students at the State Islamic University of North Sumatra (UINSU). The urgency of this research lies in the increasing dominance of TikTok in students' daily lives and how this platform shapes their behavioral and communication patterns. This study employs a qualitative method with a field research approach. The primary data sources are 10 key informants selected through purposive sampling, while secondary data are derived from books, scholarly articles, and reports. Data collection techniques include interviews, observations, documentation, and literature study. Data analysis is carried out through data reduction, data display, and conclusion drawing, with the validity of the data tested using triangulation techniques. The findings indicate that the filter bubble phenomenon has varied effects on students. On one hand, some students feel that the filter bubble limits their views on social and cultural issues, while others find it helpful in obtaining relevant and beneficial information. Additionally, TikTok also influences the language usage of students, with some reporting an expansion in their vocabulary and an increase in creativity, although it also leads to the use of more colloquial and less formal language. The discussion highlights the importance of digital literacy and critical awareness in utilizing social media. These findings align with the filter bubble theory by Eli Pariser, which emphasizes the need for exposure to diverse views to avoid intellectual isolation. Overall, the research suggests that students should be trained to use social media wisely, enabling them to maximize its benefits without falling into the shackles of the filter bubble.

Analysis of the Filter Bubble Algorithm on TikTok Concerning 'Toxic' Polarization During the 2024 Presidential Campaign

Social media is a platform for everyone to express and convey information. The emergence of social media certainly affects the behavior of each individual who uses it. This is due to a social media algorithm called the filter bubble, which can lead to polarization as in the Tiktok application and several other social media applications. Simply put, we are made to feel as if we know everything that is happening around us, when in fact we do not. This is what causes 'toxic' polarization to form an ignorant personality. In other words, 'toxic polarization' will isolate every social media user intellectually. Of course, this will be very dangerous if it happens during the 2024 presidential election campaign because it will cause many divisions. The long-term effect will certainly affect democracy so that it can cause stagnation.

Medical science in peril? analyzing the anti-vaccine rhetoric on greek facebook in the COVID-19 era

In the post truth era the limits between facts and beliefs, science and pseudo-science seem to be quite blurred. Social media platforms, such as Facebook, provide the ideal vehicle for the widespread sharing of misinformation, by fostering the creation of “filter Bubbles” and “echo chambers”. For the purposes of our study, we analyzed Facebook posts from users’ pages and public groups over a three – month period (from September 2021 to November 2021) with a primary focus on the vaccination against COVID-19. Using thematic content analysis, we attempted to explore the major themes that skeptics of COVID-19 vaccines use to justify their hesitancy. According to research findings, the anti-vaccination discourse on Facebook platform in Greece combines features such as mistrust towards politicians, the medical community, the legacy media, the establishment as a whole, and facts regarding the safety/complications of COVID-19 vaccination itself. If the medical society aspires to combat misinformation regarding health issues, then, deconstructing the anti-vaccination rhetoric will be a pivotal step for the ultimate survival of epistemological truths and scientific discourse in the public sphere.

Social Media in Shaping Public Opinion Roles and Impact: A Systematic Review

This systematic literature review examines the impact of social media on public opinion and its implications for policy-making. Utilising the PRISMA framework, the study analysed 19 articles from Scopus and Web of Science databases published between 2013-2023. The review identified five main categories of social media platforms discussed; Twitter/X, Meta (Facebook, Instagram, WhatsApp), YouTube, Chinese apps (Sina Weibo, WeChat, QQ), and several unspecified platforms. Key themes emerged across these categories, including the role of social media in knowledge dissemination, creation of filter bubbles and echo chambers, amplification of diverse voices, and spread of misinformation. The findings highlighted social media potential for real-time public opinion monitoring, facilitating engagement between policymakers and citizens, and early identification of emerging issues. However, challenges such as information credibility and algorithmic curation of content were also noted. The review suggests that strategic use of social media can raise awareness and mobilise support for global initiatives like the Sustainable Development Goals. It emphasises the need for policymakers to understand and leverage social media's influence on public sentiment while addressing associated risks. The study contributes to a nuanced understanding of how different social media platforms shape public discourse and influence policy decisions in the digital age. Future research directions are proposed to further explore the complex dynamics between social media, public opinion, and governance in an evolving technological landscape. Keywords: Social media impact, public opinion formation, digital policymaking, online political discourse, civic engagement.

Prick “filter bubbles” by enhancing consumers' novelty‐seeking: The role of personalized recommendations of unmentionable products

AbstractPersonalized recommendation algorithms inadvertently foster “filter bubbles,” wherein consumers are predominantly exposed to information that aligns with their existing preferences, limiting their exposure to novel items. This phenomenon raises ethical concerns regarding consumer well‐being, as it potentially compromises the quality of consumption decisions by reinforcing a homogeneity of information. Introducing novelty into recommendation systems is a viable strategy to counteract this issue, as the predominance of homogeneous information plays a crucial role in the formation of filter bubbles. However, there is a notable gap in the literature regarding self‐directed strategies for consumers to break through these filter bubbles. Grounded in social identification theory and utilizing a series of experimental studies, our research employs a range of analytical techniques, including ANOVA, mediation, and moderated‐mediation analysis. Our findings suggest that personalized recommendations of unmentionable products, defined as products eliciting disgust, offense, or anger due to delicacy, ethics, or fear, (vs. ordinary products) can increase consumers' novelty‐seeking by enhancing their motivation to change their implicit social labels given by intelligent recommendation systems. Nonetheless, we observe that this drive for novelty‐seeking diminishes during social‐focused recommendations because this recommendation is based on the behaviors of others in consumers' social networks rather than their actions.

Filter Bubbles and the Unfeeling: How AI for Social Media Can Foster Extremism and Polarization

Social media have undoubtedly changed our ways of living. Their presence concerns an increasing number of users (over 4,74 billion) and pervasively expands in the most diverse areas of human life. Marketing, education, news, data, and sociality are just a few of the many areas in which social media play now a central role. Recently, some attention toward the link between social media and political participation has emerged. Works in the field of artificial intelligence have already pointed out that there is a close link between the use of machine learning algorithms in social media and possible epistemic isolation, which could lead to political radicalization. The idea supporting this paper is that artificial intelligence for social media can actively put users’ deliberative capacity at risk and foster political extremism. To prove these claims, I proceed along two lines of inquiry. First, I focus on filter bubbles, namely the result of selections made by algorithms that recommend contents that meet users’ expectations and opinions. To analyze this phenomenon, I refer to the Deweyan model of experience. Second, I connect the filter bubbles problem to the Deweyan idea of deliberative and participatory democracy and Nussbaum’s concept of political compassion. The purpose of this paper is to provide a philosophical foundation that can both (1) effectively serve as a method for analyzing machine learning algorithms and their potential problems in relation to political extremism, and (2) be adopted as a standard to counter the danger of extremism associated with social media experience.