Algorithmic Nostalgia: How Machine Learning Curates Our Emotional Connections to the Past

In an era increasingly shaped by algorithmically curated media environments, nostalgia has become a powerful emotional tool within digital platforms. This study investigates how algorithmic recommendation systems on Instagram and Twitter facilitate the emergence of nostalgia-themed content and influence users' emotional engagement. The analysis explores how brand-generated throwback posts, strategically crafted to evoke collective memory and sentiment, intersect with user interaction patterns and algorithmic visibility mechanisms.Using a qualitative digital ethnography approach, this research is based on non-intrusive observation of selected public brand accounts on Instagram and Twitter. Content using hashtags such as #TBT and #ThrowbackThursday was examined to understand how digital nostalgia is amplified through algorithmic circulation. The findings reveal that algorithm-driven promotion of nostalgic content not only enhances users' emotional connectivity and sense of community but also generates concerns about data visibility and user autonomy. Some users were observed to later remove their engagements, suggesting a potential discomfort with the visibility or implications of their digital interactions.The study highlights that digital nostalgia—while fostering emotional continuity—can also reinforce the complexities of personalization and data ethics in platform economies. By focusing on the interplay between emotional experience and algorithmic mediation, the study contributes to a deeper understanding of how affective engagement and concerns over data privacy co-evolve in contemporary digital culture.