There are interesting variations in each of these two cases. Let's compare.

The person who removed the hippocampus can not make new memories. However, when he was trained to trace the stars in the mirror, the star he wrote gradually went well. In other words, he can not make memories in words, but he can make memories of skills to move his hands well. (*8)

In the Morris water maze experiment, Morris changed the position to put the rat in the aquarium every time. What happens if we put the rat in the same place every time? Even the rats with broken hippocampus had less time to reach the scaffold. In other words, the rat with broken hippocampus cannot make" memory for arriving at the scaffold from the inexperienced start," but it can make "memory for reaching the scaffold from the experienced start." (*9)

Both cases strongly suggest that memory is not one type.

In the case of human beings, it is different from the memory of words and the memory of the motor skills such as how to move the hand. For example, even people riding a bicycle are hard to explain how to ride a bicycle in words. Such knowledge that is not possible to explain in words is called "non-statement memory." (*10)

Rats do not speak words, so it is a different way to separate two types of memories.

In the Morris water maze experiment, rats can not directly observe the location of the scaffolds, and rats can observe surrounding landmarks. Ordinary rats may be creating memories of scaffold sites that can not be observed directly, based on the information of directly observable landmarks. So, even if the rat starts from a different place, it can reach the scaffold quickly.

On the other hand, broken rats may memorize the specific procedure of swimming to reach the scaffold. If the rat starts from the same place, it can reach the scaffold efficiently with the memory. However, if the rat starts from a different place, the rat needs to look for scaffold again just as if it does not learn anything.

Creating the information of a scaffolding location from the view of landmarks corresponds to the "abstraction" described in Chapter 1. Rats who were able to abstract from the concrete information (= view of the landmarks) and make the model (= the place of the scaffold) were able to answer efficiently even for new problems (= new start positions.) On the other hand, rats who remembered concrete experiences (= how to get to the scaffold from a specific start) had to do trial and error on the new problem.

I feel the hippocampus is in charge of an important phase of the learning cycle.

---

Footnotes:

*8 Also, the person did not remember that he experienced the training. His skill improves without his awareness.

*9 After learning, if you put the rat in a new place in the aquarium, normal rats quickly trace to the scaffold, whereas rats with broken hippocampus swam like they did not learn scaffold and took a long time to find the scaffold.

*10 It is difficult to explain the way to remember the act of "remembering" by words. Some experts in memory techniques seem to be able to explain how to remember, for example, "image the building in detail, leave knowledge to the place, just walk around the building when remembering." (see Method of loci) I can not explain how I recall because I can not master this memory technique.