The Algorithmic Mirror: Knowledge Creation and Self-Perception in Dating Applications

Paper Title

The Algorithmic Mirror: Knowledge Creation and Self-Perception in Dating Applications

Publication Info

• Topic area: Algorithmic mediation in dating applications and its impact on self-perception and user strategies.
• Keywords: Algorithmic mirror, dating apps, self-perception, algorithmic other, impression management, folk theories, user agency, emotional labor, algorithmic fairness, socio-technical systems.

Background and Problem

• Problem / challenge: Existing research has extensively documented the psychological impacts of algorithmic dating platforms but has not sufficiently explored the strategies users employ to interpret and navigate these systems.
• Significance: Understanding how users interact with algorithmic systems is critical for designing platforms that better support user agency and mitigate emotional labor.
• Motivation and related work: Prior studies have highlighted the commodification of self and algorithmic oppression in dating apps, but they often treat user knowledge as monolithic. This paper investigates the nuanced ways users create and share knowledge to navigate algorithmic environments, addressing gaps in human-computer interaction (HCI) and sociological research.

Solution

• Proposed approach: The study introduces the concept of the "algorithmic mirror," a recursive process where users interpret and reshape the data-driven reflection presented by dating app algorithms.
• Novelty:
  1. Development of a three-part typology of user knowledge: Folk, Personal, and Academic.
  2. Extension of sociological theories by framing the "algorithmic other" as a statistical counterpart to Mead’s "generalized other."
  3. Identification of the "dual-audience dilemma," where users perform for both human partners and algorithmic systems.
• Procedure and key techniques:
  • Conducted semi-structured interviews with 15 OkCupid users.
  • Used thematic analysis to identify patterns in user strategies and emotional responses.
  • Mapped user strategies into three knowledge types and analyzed their implications for self-perception and system design.

Results

• Concrete findings:
  • Users construct three types of knowledge—Folk (community advice), Personal (trial-and-error experimentation), and Academic (external theoretical frameworks)—to navigate algorithmic systems.
  • The algorithmic mirror profoundly impacts self-perception, leading to both emotional labor and personal growth.
  • Gendered differences in strategies were observed, with men engaging in more explicit optimization tactics due to perceived competition.
• Advantage over baselines: Provides a nuanced understanding of user strategies beyond the monolithic view of "folk theories," emphasizing proactive and generative approaches to algorithmic navigation.
• Experiments / evaluation:
  • Sample: 15 heterosexual OkCupid users aged 22–30.
  • Method: Semi-structured interviews analyzed using thematic coding.
  • Metrics: User strategies, emotional responses, and perceptions of algorithmic fairness.
• Limitations and future work:
  • Focused demographic limits generalizability to other populations (e.g., LGBTQ+ users, older adults).
  • Findings are based on self-reported experiences, which may not align with actual algorithmic behavior.
  • Future research should explore diverse populations and test design interventions informed by the "algorithmic mirror" framework.

Summary

This study investigates how users of algorithmic dating platforms create and mobilize knowledge to navigate opaque systems and how this process shapes their self-perception. Through interviews with OkCupid users, the research identifies three types of knowledge—Folk, Personal, and Academic—that users employ to interpret and adapt to algorithmic logic. The findings extend sociological theories by introducing the concept of the "algorithmic other," a statistical counterpart to Mead’s "generalized other," and highlight the "dual-audience dilemma" faced by users performing for both human partners and algorithms. The study underscores the need for transparent, negotiable, and collaborative system designs to support user agency and emotional well-being. Future work should examine these dynamics across diverse populations and algorithmic contexts.