🚀 go-pugleaf

-->What is compensating our visual system to see the slide properly in the first three cases, but not in the last unless we apply a correction to the CRT?

There is no compensation in the first three cases, because none was required, because no linear capture mechanism was involved. Film, within limits, responds to light in a similar fashion to our eyes. Film is not a linear capture mechanism.

It's not the CRT that demands the correction, it's the linear capture from the slide scanner. The fact that CRTs have a power curve that's extremely close to one that models human visual response is simply a happy accident.

The need for gamma correction in digital imaging is entirely about the difference between the way eyeballs see light, with a strong compressive nonlinearity that can be modelled quite well by a power curve of around 2.4, and the way silicon sensors see light, which is linear.