Recommendation algorithms reinforce an archetypal self instead of broadening taste

Evidence

Over the years, the Spotify algorithms have correctly identified that I tend to like “chill” music of a certain BPM: smooth, inoffensive songs from the 1960s and ’70s, or more recent ones with washy synths, echo-y guitars, and vocals that are either passive or nonexistent. As I continue to listen to the playlist, dutifully saving the songs that I like, the weekly playlist begins to hone in, if not on an archetypal song, then an archetypal mix—we could call this “the Jenny mix”—and other potential mixes are measured for their likeness to whatever the current archetype is. — Jenny Odell, How to Do Nothing, 468

Categorization

• XREF: connects to: Personalized consumption niches chain listeners to their own preferences — Mark Fisher (edited by Darren Ambrose), K-Punk: The Collected and Unpublished…, 1286; Algorithm-curated friendship narrows one’s social world into a self-reinforcing echo — Jenny Odell, How to Do Nothing: Resisting the…, 467; Algorithms reinforce existing tastes rather than challenging or expanding them — W. David Marx, Blank Space: A Cultural History of the…, 709 (+2 more)
• yield-rescued: below the book’s yield-target cutoff, included anyway for connecting to: Personalized consumption niches chain listeners to their own preferences — Mark Fisher (edited by Darren Ambrose), K-Punk: The Collected and Unpublished…, 1286; Algorithm-curated friendship narrows one’s social world into a self-reinforcing echo — Jenny Odell, How to Do Nothing: Resisting the…, 467; Algorithms reinforce existing tastes rather than challenging or expanding them — W. David Marx, Blank Space: A Cultural History of the…, 709 (+2 more)

Broader Connections

• Algorithms reinforce existing tastes rather than challenging or expanding them #corroborates (0.97) — W. David Marx, Blank Space, 709