A simple way of looking at this is to see the things that you put into the oven as either opaque reflectors of microwaves (generally good conductors), or transparent, or partly translucent, absorbing some part of the light passing through, but not all. The way a beer bottle passes visible light would be a good analogy. There also are complications when good conductors are formed into  microwave aerials, but never mind them for now. Very well, microwave-murky materials tend to absorb most of the energy within a few quarter waves (about 3 cm).  Watery, fatty, foods commonly behave like that, but with enough variation to make things tricky for a heedless cook. For example, ice only absorbs microwaves about half as well as liquid water, so it sometimes is possible to over-cook part of a frozen meal while another part is still frozen solid.  Much as an empty beer-bottle in desert sun can get painfully hot while clear glass barely gets warm, the food acts as a lossy medium and absorbs energy. That is why we cook with MWOs. The ice stays cooler for longer. Most everyday plastics are MW-transparent, and so are MO-safe ceramics, which is why we can  use them for cooking. However, some ceramics, especially those rich in ferrites, are nearly opaque. Some of them absorb MWs within a few mm, getting quite hot comparatively abruptly. Obviously any soup behind them gets shaded and takes longer than usual to warm up.  Lately MWO ceramic cookware on the market has been produced with ferrite-rich panels built in to cook the food in much the way that a stove hotplate might do. However, many ordinary ceramics, especially some of the brown materials, contain a little ferrite or other MW-absorbant impurities.  They are not intended for MW cooking, but they can get very hot too. Presumably yours was such a bowl. Reflective metal bowls either act as aerials with concentrated local effects, or simply reflect the MWs, hardly heating their contents at all. Happy souping!   Jon