Cool, thanks!!

Interesting, do you have a paper on this?

Analyzing 1M Spotify users, we show how “wide social influence” exposes people to content beyond their usual repertoires—decoupling collective outcomes from initial preferences. The key: partial taste overlap between senders and receivers. Too little similarity, no influence; too much, no novelty.

...and not way earlier?

Interesting, but what is the explanatory mechanism here? If the reasons for the far right shiftgo that far back, why did the shift happened more or less recently?

Spannend! Will ich lesen!