You are right, the individual propensity to go to the office is going to be highly sensitive to the number of people already going to the office. I believe this

You are right, the individual propensity to go to the office is going to be highly sensitive to the number of people already going to the office. I believe this to have a steep s-curve shape; with little number of people at the office there is little incentive to show up. But after a critical threshold of office-goers, it will very rapidly be a career/productivity disadvantage to opt not showing up. Only thing that can change this curve is active office occupancy ceilings enforced by the company. Which kinda sucks, my employer might be good at writing web services, but how much do I trust them doing good epidemiology?