4.4. Testing the idea of inflation
An extremely useful sentence to bear in mind in considering how to test the inflationary paradigm is the following:
The simplest models of inflation predict a spatially-flat Universe containing gaussian distributed tensor and adiabatic scalar perturbations, which are in their growing mode with almost power-law spectra.The underlined phrases indicate the characteristic predictions, while the emphasised word `simplest' stresses that these predictions are not all generic. With the possible exception of the power-law form, the class of simplest models encompasses all the single-field models, plus many other models which prove to be dynamically equivalent to them
So far, all these characteristic predictions has been borne out, though the strength of the tests differs [16]. Only spatial flatness has really been confirmed at a convincing level, and even that only very recently [1, 2]. The others are promising hypotheses which are part of the currently-favoured view as to how structure arose.
The immediate goal is to test these hypotheses, and if they remain valid to use measured quantities such as n and r to establish which subset of the simplest models best fits the data. Our current aim, therefore, is to test the simplest models of inflation. If they are found wanting, then the way in which they fail will be indicative of whether it is worthwhile to study the wider range of inflation models, or if attention is best refocussed elsewhere. One should also stress that the aim is to test inflation as the sole origin of structure; one can consider admixtures of the inflationary perturbations with those of another source; this is fine if positive evidence is forthcoming but it will of course be impossible ever to exclude the possibility of an admixture at some level.