Analyzing Privacy Dynamics within Groups using Gamified Auctions
Hüseyin Aydın*, Onuralp Ulusoy*, Ilaria Liccardi, Pınar Yolum · 2025
Read Analyzing Privacy Dynamics within Groups using Gamified Auctions by Hüseyin Aydın*, Onuralp Ulusoy*, Ilaria Liccardi, Pınar Yolum (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-2502-10788.txt.
Introduction
Hüseyin Aydın, Onuralp Ulusoy, Ilaria Liccardi, and Pınar Yolum study group privacy choices in a 2025 computing paper. Shared items like group photos hold facts about many people at once. Each person in a photo may want a different sharing choice. Some systems settle such clashes with auctions, where members bid to get their way.
The team built a game, called RESOLVE, to watch how real users act in these auctions. Players' sense of their own privacy differed sharply from their sense of group privacy. Many fought for their own choice even when it cost others. At other times, players teamed up to guard someone else's privacy. The results should shape fairer tools for shared content.
Catalog record
| CREATOR | Hüseyin Aydın*, Onuralp Ulusoy*, Ilaria Liccardi, Pınar Yolum |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 136 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2502-10788
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Analyzing Privacy Dynamics within Groups using Gamified Auctions Hüseyin Aydın*, Onuralp Ulusoy*, Ilaria Liccardi, Pınar Yolum Abstract . Online shared content, such as group pictures, often contains information about multiple users. Developing technical solutions to manage the privacy of such "co-owned" content is challenging because each co-owner may have different preferences. Recent technical approaches advocate group-decision mechanisms, including auctions, to decide as how best to resolve these differences. However, it is not clear if users would participate in such mechanisms and if they do, whether they would act altruistically. Understanding the privacy dynamics is crucial to develop effective mechanisms for privacy-respecting collaborative systems. Accordingly, this work develops RESOLVE, a privacy auction game to understand the sharing behavior of users in groups. Our results of users’ playing the game show that i) the users’ understanding of individual vs. group privacy differs significantly; ii) often users fight for their preferences even at the cost of others’ privacy; and iii) at times users collaborate to fight for the privacy of others. CCS: Security and privacy Social aspects of security and privacy CCS: Security and privacy Usability in security and privacy 1. Introduction Privacy is often understood as a right of individuals to choose what information they want to share about themselves ( Acquisti et al., 2015 ) . Typical way of treating privacy online is through information consent. In essence, this paradigm puts the individual in charge to declare what information they would like to share, with whom, and for what purposes. General Data Protection Regulation (GDPR) is a prime example of this paradigm, where users are informed about the content that is being accessed about them ( Commision, 2018 ) . The user then gives an informed consent as to how their personal data will be shared. As a result, with appropriate techniques, it is possible to detect whether personal data have been leaked without the person’s consent or used against their will. Many times our content is not simply personal data, but pertains to multiple users. Consider a group picture that is posted on social media: There is a single person that uploads the picture, whom might be interpreted as the owner of the content. However, the decision of sharing this content will have privacy implications on the others that are in the picture as well as the uploader. Similarly, in a collaborative editing environment, two individuals might edit the same document and then one of the editors can choose to share the document with others, without the need to consult the other person, again possibly creating privacy violations. Put simply, these happen because the content that exists on collaborative systems, such as co-edited documents or group pictures, does not have to a single owner, but multiple co-owners , with possibly different privacy preferences ( Such and Rovatsos, 2016 ; Hu et al., 2012 ; Sleeper et al., 2013 ) . Hence, it is not appropriate to assume that the sharing party is the only owner of the content and that her privacy settings should apply to the content. Protecting the privacy of co-owned content requires a shift from personal privacy to multi-party privacy. Since each user of online content might have different privacy preferences, solutions that aggregate the individual preferences indiscriminately and automatically result in privacy violations of some co-owners ( Squicciarini et al., 2018 ; Kokciyan et al., 2016 ) . Ideally, systems should employ group-decision mechanisms that enable each user to express her opinion as to whether a content should be shared and, if so, how. There have been recent technical approaches to create such mechanisms ( Such and Rovatsos, 2016 ; Squicciarini et al., 2009 ; Kökciyan et al., 2017 ; Ulusoy and Yolum, 2021b ) without any studies as to whether users would employ and participate in such mechanisms. More importantly, how differently would they behave in a multi-party situation through this type of a mechanism? What would be the factors behind those behaviors? Considering such privacy dynamics of users in groups leads us to these particular research questions: RQ1: How do people determine their content sharing preferences in a multi-party situation? RQ2: When do people tend to keep their privacy preferences, especially even if they contradict the majority’s wishes? RQ3: Which circumstances lead people to change their privacy preferences? This paper addresses these questions by conducting an empirical study on group images. To realize this study, we have created a novel game RESOLVE that uses a group-decision making system based on auctions, where each player bids on how much they would like to see a group picture shared or not shared. The group pictures are created using the participant’s own pictures to make them realistic. By creating a systematic quiz for the game, we ensured that the participants understood the rules and workings of the game well. For the purpose of this study, we have one player as participant of the study and three other players that are software agents (configured by us). This enables us to create interesting situations, such as only the participant would not like to share a group picture while the other three would. Unique to this game is the possibility to revisit privacy choices on the same picture after seeing the preferences of others. By enabling a participant to make a privacy choice before seeing others’ choices and afterward, we analyze whether the participants’ perception of privacy changes based on the choices of others and whether participants fight for their privacy in such situations. We observe that participants’ privacy declarations on individual pictures could change when the individual picture is merged into a group picture. Moreover, some participants actually fight for their privacy even when they know they are in the minority. On the other hand, we also see some participants ‘switching sides’ when their choice is not in the majority, to help others preserve their privacy. These and other observations we describe later are important insights for developing the next-generation software tools to manage the privacy of co-owned data, both on social media as well as in collaborative tools. Our major contributions are the following: • We developed a novel game RESOLVE . This game is configurable to test user behaviors in various multi-party privacy situations with group pictures of interest. • We developed material to teach how to play RESOLVE as well as a systematic quiz to evaluate and help a user play the game. • We have performed a user study with 40 participants to evaluate privacy dynamics of users while shifting from individual pictures to group contents as well as the effect of knowing the preferences of others. The rest of the paper is organized as follows. Section 2 summarizes the necessary background information and related work for the study. Section 3 presents the user study including all the features of our game RESOLVE and the demographics of the participants. User profiles based on their interactions with the game are provided in Section 4 . Section 5 examines the observed behaviors through the game. Before concluding the paper, the validity and limitations of the study are discussed in Section 6 . 2. Background & Related Work Different terms are used for the privacy decisions involving more