It has often been assumed that one or the other of these two types of representations underlies cognitive processing in a certain domain of cognition. For instance, in research on thinking, memory, and language processing, the traditional assumption is that properties, objects, situations, and events are captured by means of amodal representations. These representations typically abstract from the detailed aspects of the specific state of affairs that is being represented. For instance, the meaning representation of a word such as “dog” will include symbols for typical features of dogs whereas in research on perception, it is often assumed that the relevant representations are modal in nature. When perceiving, for instance, a dog, it is assumed that humans create a rather specific representation that preserves many properties of this dog.
