Paper Type

ERF

Description

We consider the problem of how a platform designer, owner, or operator can improve the design and operation of a digital platform by leveraging a computational cognitive model that represents users’ folk theories about the platform as a sociotechnical system. We do so in the context of Reddit, a social-media platform whose owners and administrators make extensive use of shadowbanning, a non-transparent content moderation mechanism that filters a user’s posts and comments so that they cannot be seen by fellow community members or the public. After demonstrating that the design and operation of Reddit have led to an abundance of spurious first-party suspicions of shadowbanning in cases where the mechanism was not in fact invoked, we develop a computational cognitive model of users’ folk theories about the antecedents and consequences of shadowbanning that predicts when users will attribute their on-platform observations to a shadowban. The model is then used to evaluate the capacity of interventions available to a platform designer, owner, and operator to reduce the incidence of these false suspicions. We conclude by considering the implications of this approach for the design and operation of digital platforms at large.

Paper Number

1961

Comments

SIG CORE

Share

COinS
 
Aug 10th, 12:00 AM

The Design and Operation of Digital Platforms under Folk Theories of Sociotechnical Systems

We consider the problem of how a platform designer, owner, or operator can improve the design and operation of a digital platform by leveraging a computational cognitive model that represents users’ folk theories about the platform as a sociotechnical system. We do so in the context of Reddit, a social-media platform whose owners and administrators make extensive use of shadowbanning, a non-transparent content moderation mechanism that filters a user’s posts and comments so that they cannot be seen by fellow community members or the public. After demonstrating that the design and operation of Reddit have led to an abundance of spurious first-party suspicions of shadowbanning in cases where the mechanism was not in fact invoked, we develop a computational cognitive model of users’ folk theories about the antecedents and consequences of shadowbanning that predicts when users will attribute their on-platform observations to a shadowban. The model is then used to evaluate the capacity of interventions available to a platform designer, owner, and operator to reduce the incidence of these false suspicions. We conclude by considering the implications of this approach for the design and operation of digital platforms at large.

When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.