Another missing bias: people tend to think of wealth in terms of income rather than accumulation of assets. That is, they think in terms of what comes in over the short term rather than what's built up over the long term.

To clarify: imagine a society where everyone makes the exact same income up until retirement (50 years in the workforce), and everyone saves the same percentage of their income each year in a personal retirement fund. Assume also that retirees spend their assets down to zero by death in a linear way. In this society with perfectly even income distribution, wealth by quintile will be as follows:

  Top quintile (40+ years working, or newly retired): 36%
  Second quintile: 28%
  Middle quintile: 20%
  Fourth quintile: 12%
  Bottom quintile (<10 years working): 4%

According to the article, the group "most heavily skewed toward a top-heavy distribution" (those earning >$100k) gave their ideal as only slightly out of line with this. I don't think it's rational to conclude the "most skewed" group actually believes wealth should be distributed according to an almost-identical-salaries-for-all model; I think it's more rational to conclude that the people surveyed are using a flawed mental model of wealth that doesn't properly account for accumulation over time.