Yes, you would think an LCD pixel simply changes color as instructed after a finite delay. But perhaps the pixels as a group are controlled in a fashion that has the effect of mimicking a CRT scan? I know, Google...

Interlacing just scans every other line in one pass and then on the second pass it scans the other lines. All lines are scanned, but takes two passes for everything to be scanned. Any effect of lines between is just because the lines from the previous pass are starting to fade in comparison to the freshly scanned lines. BTW, only CRTs and ALiS plasmas are able do interlaced.

Two fields (odd and even) of an interlaced image combine to make a full frame. Persistence of vision is used on a CRT to blend the images together, and on an LCD the image is deinterlaced digitally before it's displayed (i.e. it waits until it has received both fields and displays them both combined.)

If you can see blank lines, then your display is crappy.

Ever noticed that CRT TVs often have this "slot mask" phosphor layout?

I always figured that this was designed to work with interlaced video, since it blends the half frames together in both horizontal and vertical direction. But I can't find any conclusive information on it.

An interlaced LCD produces combing artifacts instead of scanlines. Take the PSP 3000 for example. It just looks like lines are shifted or jitter.

My primary gaming displays are 1920x1080. One is a 1080i HDTV (60fps/30fps effective interlaced,) one is a 1080p (60fps) computer LCD.

The HDTV is run by a Core 2 Quad QX6700 with a Radeon 5770, the computer LCD is on a Core 2 Duo T9500 with Radeon 2400XT (which also has a 1680x1050 display, yup, it's an old iMac.)