Lapidus said:
In the pop-science books I read both.
Pop science books are not good sources if you actually want to learn about the science.
For a quick look at a simple version of the physics behind inflation, check out Figure 10 in section 6.1 of this series of lectures (the series itself is worth reading):
http://arxiv.org/abs/0907.5424
Basically, the scalar field ##\phi## starts out on the flat area on the left. While it is there, it can drive inflation (because, as the text says, its potential energy exceeds its kinetic energy). However, since that area is not perfectly flat, the field will sooner or later start "rolling" to the right down the slope. The slope starts out gentle (this is the "slow roll" talked about in the text), but then grows steeper. When it gets steep enough, the scalar field can no longer drive inflation (actually this is a combination of increasing steepness and time spent rolling--the key criterion is that the field's kinetic energy now exceeds its potential energy), so inflation stops (this is the point marked ##\phi_{end}##). The field ##\phi## then ends up in the trough marked "reheating"; once it's there, it can't get out, and its kinetic energy (from rolling down the slope) gets converted to ordinary matter and radiation (this is the "reheating"), which is expanding rapidly and starts the standard hot Big Bang process.
Of course, the model described above is not necessarily complete. In this model, the scalar field is always there, and the starting state in which it is up on the flat area on the left is called the "false vacuum" state. The "phase transition" is the process of the field rolling down the slope, which eventually stops inflation; there is no phase transition needed to start it, since it's already happening at the start of this model (when the field is up on the flat area on the left). The end state of the model, with the scalar field down in the trough, is called the "true vacuum" state. (This is how I was using those terms in my previous post.) But the model evidently does not address the question of how the scalar field got up on the flat area in the first place. Was it always there? Or did some previous process put it there?
I'm unable to find a good online source discussing the details of Guth's original model, but IIRC it did include a previous transition from a state he was calling "false vacuum" to the state described above, with the scalar field on the flat area on the left--which is
also called "false vacuum" by many sources, as I noted above. This sort of confusing terminology is one reason why pop science books are not good sources: they virtually never explain their terminology or address the fact that there are multiple possible meanings of terms like "false vacuum".