The idea was that a single player (usually the human, but the algorithm applies to AI players as well) could not get too far ahead in its research without incurring an increasing penalty. I guess they used a KeyCiv pegged to power rating so that it changes as your power changes, masking it a bit but also giving it some flexibility as different players will be at different levels.
In terms of reality simulation, you could look at it as increasing resistance toward the technical advances of a single civ that is getting way ahead, since a lot of discoveries and research is the result of international sharing and peer review. More importantly, in terms of game mechanics, it rebalances the game and works against one civ leaving the others in its dust.
There was some sort of interview I saw a while back with either Sid or Brian where they explained the reasoning behind it. I've lost the link I found that at, but it should be out there somewhere. Check over at Apolyton.net too.
In terms of reality simulation, you could look at it as increasing resistance toward the technical advances of a single civ that is getting way ahead, since a lot of discoveries and research is the result of international sharing and peer review. More importantly, in terms of game mechanics, it rebalances the game and works against one civ leaving the others in its dust.
There was some sort of interview I saw a while back with either Sid or Brian where they explained the reasoning behind it. I've lost the link I found that at, but it should be out there somewhere. Check over at Apolyton.net too.