Influence of the majority group on individual judgments in online spontaneous conversations.

Diletta Goglia a , Davide Vega a and Alessio Gandelli a · 2025

Read Influence of the majority group on individual judgments in online spontaneous conversations. by Diletta Goglia a , Davide Vega a and Alessio Gandelli a (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-21092.txt.

Introduction

Online threads about everyday dilemmas reveal how crowds bend individual choices, sometimes in a surprising direction. Diletta Goglia, Davide Vega, and Alessio Gandelli study that influence in a 2025 paper accepted by EPJ Data Science. Their setting is anonymous, spontaneous social media conversation.

The authors measure disagreement within each conversation from digital traces of judgments. Bayesian regression then tracks how judgments form before and after exposure to the group's stance. A linguistic analysis follows changes in how people express and justify their views. The results show anti-conformity: individuals keep the majority's positive or negative orientation, yet diverge from its exact stance. The finding refines classic conformity theory for the age of anonymous online discussion.

Catalog record

CREATORDiletta Goglia a , Davide Vega a and Alessio Gandelli a
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE88 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2509-21092

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Influence of the majority group on individual judgments in online spontaneous conversations. Diletta Goglia a , Davide Vega a and Alessio Gandelli a Abstract This study investigates how the majority group influences individual judgment formation and expression in anonymous, spontaneous online conversations. Drawing on theories of social conformity and anti-conformity, we analyze everyday dilemmas discussed on social media. First, using digital traces to operationalize judgments, we measure the conversations’ disagreement and apply Bayesian regression to capture shifts of judgments formation before and after the group’s exposure. Then we analyze changes in judgment expression with a linguistic analysis of the motivations associated with each judgment. Results show systematic anti-conformity behaviors: individuals preserve the majority’s positive or negative orientation of judgments but diverge from its stance, with persuasive language increasing post-disclosure. Our findings highlight how online environments reshape social influence compared to offline contexts. keywords Reddit, Opinion dynamics, Online social media, Communication norms, Social conformity † † affiliation: a InfoLab, Department of Information Technology, Uppsala University, Sweden 1 Introduction Social norms delineate acceptable beliefs and behaviors within communities, fundamentally shaping social structures and human interactions ( Shifman et al. (2025) ) . These norms guide both individuals’ judgment formation , influencing how they interpret and evaluate circumstances, and individuals’ judgment expression , determining when they feel comfortable or compelled to declare their judgments publicly ( Bursztyn et al. (2020) ) . For example, in political discussions, people may refrain from expressing dissent when they perceive the position of a group as strongly opposed to their own ( Guo et al. (2023) ) . In online environments, social norms are often blurry and ambiguous, and their interpretation is challenging due to limited contextual cues and anonymity ( De Candia et al. (2022) ) . Users often misperceive the predominant opinion, resulting in social phenomena like pluralistic ignorance and majority illusion ( Scheper & Bruns (2025) ) . Pluralistic ignorance emerges when individuals privately reject a norm or opinion but assume (incorrectly) that others accept it, hence going along with it publicly ( Bicchieri & Fukui (1999) ) . Majority illusion refers to the individual perception of a particular norm or opinion to be more common than it actually is, often because it is held by highly visible users ( Lerman et al. (2016) ) . Despite these difficulties, individuals observe other users online and adjust their actions and thoughts accordingly, rationally and collectively constructing new shared rules and expectations. For this reason, digital platforms thrive the change of existing social norms and the creation of new ones. Within such digital ecosystem, the majority of conversations are spontaneous , meaning that they emerge from users’ intrinsic motivations to share and engage (rather than responding to formal inquiries such as surveys), and evolve organically without external prompting or structured guidance. Spontaneous conversations encompass playful jokes, nonsenses, and discussions that span an open-ended spectrum of topics, from everyday life situations to complex social issues. For instance, a recent phenomenon spreading wildly across TikTok and Reddit is the “Italian brainrot” ( Jagun (2025) ) , consisting of AI‑generated pictures of creatures, which generate a considerable amount of spontaneous discussions. Despite the increasing availability of spontaneous conversations, a considerable portion of the literature in computational social science studied human judgments formation and expression focusing on three main approaches. The first consists on the use of evidence collected from lab experiments and surveys ( Das et al. (2014) ; Cinnirella & Green (2007) ; Kyrlitsias et al. (2020) ; Shin (2025) ; K.K. Kim et al. (2019) ) . The second approach grounds the analysis direct interaction between single individuals ( Eagly & Chaiken (1993) ; Griffin (2006) ; M. Sherif & Hovland (1961) ; C.W. Sherif et al. (1965) ) . The third approach narrows experiments to limited subject areas, such as marketing and political debates (e.g., product recommendation, elections, propaganda speech, climate change; ( Novotná et al. (2023) ; Amelkin et al. (2019) ; Messaoudi et al. (2022) ; Cortis & Davis (2021) ; Widmann & Simonsen (2025) ; M.A. Taylor (2024) ) ). A missing aspect in the literature is the impact of an unidentified group of users on individual judgments in online spontaneous discussions, a crucial social communication aspect to explain phenomena like collective actions ( Greve et al. (2016) ) . In this work we measure whether and to what extent individuals align their judgments with the majority group in online spontaneous conversations. We contribute to the understanding of how the exposure to an hidden online group influences individuals’ judgment formation (what they think) and expression (how they say it). Hence, our research question is the following: RQ: Is the reveal of the majority (group) judgment affecting the expression of individual judgments in anonymous and spontaneous online discussions? We draw on established work in cognitive and social psychology concerning collective conformity (Section 2 ) and we investigate the applicability of offline socially grounded theories to online interactions. We consider the two following competing hypotheses: H1: When the group judgment is publicly disclosed, individual judgments collectively converge towards it. Social norms of collective conformity hold for online, anonymous, and spontaneous conversations. H2: When the group judgment is publicly disclosed, individual judgments collectively diverge from it. The online scenario, favoring disinhibition, exerts a strong influence on individual behavior thus confirming the occurrence of anti-conformity norms. We study the group influence on individuals in spontaneous and anonymous conversations on social media platforms, focusing our attention on discussions around everyday dilemmas (Section 3.1 ). We start with measuring the individual exposure to the group by leveraging digital traces provided by the platform. By exploiting this information as the ground truth of both individual and majority judgment, we bypass any uncertainty, ambiguity, or approximation in our measurement (Section 3.2 ). Then, we measure the change of both individual judgments’ formation and expression after the exposure to the majority group. To compare judgments’ formation before and after the group exposure, we compute the disagreement of a discussion as the multi-label entropy of the judgments expressed before and after. This measure represents a proxy for conformity effect (Section 3.3 ). To quantify the collective influence exerted on individual judgments, we employ a Bayesian multivariate regression approach (Section 3.4 ), conducting the analysis at the group level. Subsequently, in order to examine judgments’ expression , we conduct a linguistic analysis on the motivations around individual judgments, expressed in the text (Section 3.5 ). Finally, we assess the discrepancy in such expression following exposure to group influence. Our results show that: • The exposure to the group judgment has a notable impact on individual judgments’ formation (Section 4.1 ). Users exhibit a systematic tendency to not conform their judgment to the majority group (Section 4.2 ). Following the exposure, individual judgments preserve the group’s positive or negative orientation, although they differ from it (Section 4.3 ). Furthermore, judgments directed to the main character of the story discussed in the conversation exhibit the strongest anti-conformity effect towards the majority. • The public disclosure of the majority judgment impacts individual judgments expression as well. Regardless of what the majority is, after its disclosure, users continue to engage in discussion while expressing fewer judgments (Section 4.3

Read the complete text