When the Algorithm Understands You (or Doesn’t): LGBTQ+ Youth’s Identity-Related Interactions with Algorithmic Media

During their daily media use, youth interact with algorithms. Drawing on the algorithm responsiveness process framework, this paper examines the co-construction of identity between youth and algorithmic media, focusing on perceived algorithm responsiveness (PAR), perceived algorithm insensitivity (PAI), and algorithm training as sociotechnical processes that structure how LGBTQ+ youth perceive themselves. Study 1 ( N = 335), using six weekly diaries among LGBTQ+ youth, showed that affirming LGBTQ+ representations curated by algorithms predicted higher PAR and lower PAI, especially when less algorithm training was needed. In turn, higher PAR and, unexpectedly, higher PAI predicted greater identity pride. Study 2 ( N = 664), a cross-sectional survey, found that LGBTQ+ youth reported lower PAR than non-LGBTQ+ youth. An indirect effect of algorithm training on self-concept clarity via PAR emerged for non-LGBTQ+ participants but not for LGBTQ+ participants. These findings highlight social identity in algorithmic media as a relational, dynamic process, and LGBTQ+ youth can shape content recommendations for identity-related needs.