The diagrams represent light coming out of the screen towards us. There are therefore oscillating electric and magnetic fields in the plane of the screen. We don't need to consider the magnetic field, so I've left it out. I'm assuming (to make things easy) that linearly polarised light from one slit has an electric field increasing in the upward direction while light from the other has an electric field increasing to the right. Hence the black arrows. [Soon the upward and right-ward fields will reduce and then become downwards and leftwards respectively, and so on. In other words the fields are oscillating.]
Polaroid filters let through only the electric field components oscillating only in one directionalignment. As you can see from the diagrams, the components of these fields that get through the analyser (eraser) polaroid will be in phase with each other when this polaroid is at +45°, but in anti phase when it is at –45°. The diagram shows the electric fields at a particular time, but these phase relationships continue to be true at other times.
So an extra half wavelength (or half a wavelength less) of path difference will be needed to give the same sort of fringe (bright say). Therefore there will be dark fringes where there were bright fringes previously, and vice versa.