Narrow-line Seyfert 1 galaxies are defined on the basis of their line widths, and they are generally considered to be high Eddington ratio sources. But in the context of the Quasar Main Sequence, high Eddington rate sources are those which have weak [O III] lines and strong Fe II lines. There is an overlap between the two populations, but they are not identical. Thus these two selection criteria give a different view on which objects are actually high Eddington ratio sources. We discuss this issue in the context of the broad band spectral energy density, emission line shape modeling, Fe II pseudo-continuum strength, and the level of X-ray variability. We also discuss the issue of the viewing angle and the insight one can gain from spectropolarimetric observations. We conclude that there is no single driver behind the Quasar Main Sequence, and the Eddington ratio of the source cannot be determined from the source location on the optical plane alone. On the other hand, an expected range of AGN parameters combined with a simple model of the Fe II production represent well the observed coverage pattern of the plane, with not much effect needed from the dispersion due to the viewing angle.