Breaking away from the anti-social-media consensus that's bubbled up here, I haven't seen any discussion about how much the baser human psyche craves tribalism

Breaking away from the anti-social-media consensus that's bubbled up here, I haven't seen any discussion about how much the baser human psyche craves tribalism & self-grouping more than anything else. https://arxiv.org/pdf/2305.16941.pdf https://arxiv.org/pdf/2305.16941.pdf (page 5) As shown in the paper itself, regardless of the political tribe, there's a persistent positive feeling that comes with reinforcing the viewpoints of the in-group. (page 4) It should also be noted that the personalized tweets are self-reported to be rated slightly higher than the chronological tweets, with political tweets being the exception. In short, when politics is filtered out, the Algorithm is doing its job well. > 1.3 Tweet value to reader > For each tweet, we also have users assess whether they value seeing that tweet or not (“When you use Twitter, do you want to be shown tweets like [@author-handle]’s?”, where users choose between “Yes”, “No”, and “Indifferent”). We find that overall, tweets shown by the algorithm are rated slightly higher (0.05 SD, p = 0.024). Interestingly, however, when restricting to political tweets, those recommended by the personalized algorithm lead to significantly lower user value than those in the chronological timeline (−0.14 SD, p = 0.004). Dataset-wise, it's heavily skewed towards left-&-far-left Democrats (page 12, Tables 5 & 6), & young people (page 13, Table 4).