When we read a text, we pick up visual cues based on font size and clarity,
the presence or absence of "pictures," spelling, syntax, discourse cues, and topic.
In other words, we integrate data from a text including its smallest and most
discrete features as well as its largest, most abstract features. Usually, we
don't even know we're integrating data from all these levels. In addition, data
from the text is being integrated with what we already know from our experience
in the world about all fonts, pictures, spelling, syntax, discourse, and the topic
more generally. No wonder reading is so complex!