1. The creation of theories
One of the things I love about reading is the way in which you have to generate theories about what the text means in order to understand it. This is very often a subconscious process but the act of understanding is one in which you create an image in your mind of what the text is saying and compare that image to what the text says. You experience confusion when the text says something that contradicts the image you have created. This confusion is not necessarily bad but acts as a trigger to create a new theory/image through which to view the text.
In this sense reading is not a passive activity despite how it may seem so. It is a conversation in which you keep asking the text “Do you mean this?” and assume a yes till something contrary comes up.
Nuances of meaning.
I love it when you see a word in a different context – or used in a different way – to how you are used to seeing it. This is one of the reasons why I enjoy reading a lot of classical literature because as a result of the change language undergoes over time you are more likely to see language used in a foreign way by reading older texts than you are by reading contemporary texts.
One example of this is the word “impression” as used in 17th/18th century philosophical texts. Before reading such texts I had only seen this word used in a social sense. The meaning being something like the image of yourself you imprint upon another person (that rendition of the contemporary usage of impression is heavily influenced by the meaning given in philosophical texts). But in philosophical texts it has a more technical meaning as that which the world does to the senses. Though the meaning of impression as an im-press, as a stamp onto something, seems obvious in retrospect I never made that link till I read “Inquiry concerning human understanding”.
The reason this is a particularly beautiful result of reading is that words are not thoughts. There is a pre-lingual space in which thoughts occur and are translated from this space into words (which is why it’s possible to say the same thing in the same language in many different ways and the same thing in different languages).
In a sense you can see thoughts as ideas like triangle and the verbalizations of thoughts as specific triangles. Just as an isosceles gives rise to different characteristics to a right-angle triangle so different verbalizations (renderings if you will) of the same thought can give rise to different associations. So seeing the same word in many dissimilar instances enables one to use that word to render more concepts in more ways which in turn enables one to make new associative networks!