Kosslyn Computational Model

Kosslyn’s model goes and meets Paivio’s model in many ways. According to Kosslyn, our mind possessess two permanent information storages: images and propositions.

Via the images storage you will find information about the object’s form (through images) and through the propositional one, you’ll find the object’s meaning. These two types of representation are separated, however, they must join up in order for us to be able to represent the object in our minds.

For Kosslyn, images are represented in a spacial medium, with a limited capacity. It provides us the ability to deal with spatial relationships and object rotation, but the information depends on some perspectives.

For example, in a 1975 study, Kosslyn asked the subjects to imagine a rabbit and an elephant. Then, he asked them to imagine a rabbit and a fly. The rabbit had way more details in the second condition, because it’s the larger animal and our representation acts accordingly to that fact.

Two levels of image representation

According to Kosslyn, the image representation has two components, a superficial representation and a deep one.

The deeper representation is the stored information in Long-term Memory, from which the superficial representations branch out. These ones happen at visual level in the active memory.

In deep representation, the images could be subdivided in literal or perceptive appearance of the object and the list of propositions describing the object.

In conclusion, looking at a practical example: an image of the country’s president implies both the existence of an image and a proposition, in this case the presidency.