Guards posts can be pretty, but the general concensus is that guard posts are largely a joke in this game. You're almost always better off spending the money on units. Maybe early on in the game, if you get the hp upgrade, and combat chasis don't advance too quickly (or weapons), then maybe GPs are useful.
Take a look at their hitpoints though. They just don't compare to that of a tiger. Plus, the tiger has a dual turret, and GPs will only ever have single turrets. By late game, a single GP is so far outpowered by a single tiger that there is just no point in building them. The only way to properly defend your base is with units. Plus, it's much easier and faster to build and deploy units than a guard post. Having to deal with building structure kits, loading convecs, deploying the GP, and returning the convec to the SF is a lot more than building a unit and moving it where you want it. Especially considering how much micromanagement this game makes you do already.
Yes, I love GPs, they can make you base beautiful, and are strangely satisfying to build a nice wall of, but they just aren't worth it when you think of what you're getting out of the time and money spent on them. Go for units if you're going to find yourself in a fight.
Another reason why not to go for defense, is in a Last One Standing type of scenario, you simply can't win unless you attack. You can't destroy their base if you never leave yours. The best you can hope for is a draw, and since those don't happen, you pretty much resign yourself to defeat. That's not to say waste your units on pointless attacks, but rather plan on attacking.
I'm not sure if lights do anything to fool the computer. But hey, it can be fun practice.
Oh, and additional reasons for not splitting scientists on research.... Little know fact, but the number of scientists you have assigned to a research task affects the efficiency of each scientist. It turns out that the more scientists you have assigned to a task, the more efficient each scientist will be. Here's a table I pulled out of the exe on how it works.
Edit: The numbers to the left are the max scientists you can assign to the research, and the numbers along the top are the number of assigned scientists.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
4: 7 9 11 13
5: 7 9 11 12 13
6: 7 9 10 11 12 13
7: 7 8 9 10 11 12 13
8: 7 8 9 10 11 12 12 13
9: 7 8 9 10 11 11 12 12 13
10: 7 8 9 10 11 11 12 12 12 13
11: 7 8 9 10 10 11 11 12 12 12 13
12: 7 8 9 10 10 11 11 11 12 12 12 13
13: 7 8 9 10 10 11 11 11 12 12 12 12 13
14: 7 8 9 10 10 11 11 11 11 12 12 12 12 13
15: 7 8 9 10 10 10 11 11 11 11 12 12 12 12 13
16: 7 8 9 9 10 10 10 11 11 11 11 12 12 12 12 13
17: 7 8 9 9 10 10 10 10 11 11 11 11 12 12 12 12 13
18: 7 8 9 9 9 10 10 10 10 11 11 11 11 12 12 12 12 13
As you can see, that's about 150% the research rate from each scientist if you've assigned enough to the task, as opposed to only 1.
Edit: Updated table. It was missing the last entry of each row.