Small (1992) mentions a side-effect of dividind text displays into visual chunks. I provide indirect corroboration from techniques of memorizing texts in Classical Antiquity and the Middle Ages, according to which it is easier to recall text by remembering the way it appeared in the "original" than by memorizing it as an isolated string.
2. Quintilian (184.108.40.206-33), writing in the first century A.D., recommends that:
There is one thing which will be of assistance to everyone, namely, to learn a passage by heart from the same tablets on which he has committed it to writing. For he will have certain tracks to guide him in his pursuit of memory, and the mind's eye will be fixed not merely on the pages on which the words were written, but on individual lines, and at times he will speak as though he were reading aloud. Further, if the writing should be interrupted by some erasure, addition or alteration, there are certain symbols available, the sight of which will prevent us from wandering from off the track. This device bears some resemblance to the mnemonic system which I mentioned above [the "architectural" system of loci], but if my experience is worth anything, it is at once more expeditious and more effective. 
This advice is an application of the idea that recall works best if you recreate the context in which you first experienced and, in this case, first memorized something -- a phenomenon well known to psychologists (for example, Baddeley 1990, 268-271; and Neisser 1988).
3. Quintilian (11.2.28-29) further suggests:
If certain portions prove especially difficult to remember, it will be found advantageous to indicate them by certain marks, the remembrance of which will refresh and stimulate the memory. For there can be but few whose memory is so barren that they will fail to recognize the symbols with which they have marked different passages.
In other words, if your text does not come ready-made with signs of division or markings, you should feel free to put them in. While today we sometimes frown on marking the printed page, ancient texts were supposed to be "annotated" by the owner. The classical habit of writing without any spaces or punctuation (scriptura continua), in fact, forced all readers to punctuate the text for themselves either mentally or physically with some kind of marking. If someone else punctuates the text, in effect does all the work for you, you will not be able to remember that text as easily. Mary Carruthers (1990, 247), in her study of memory in the Middle Ages, convincingly argues that in medieval manuscripts "the basic function of all page decoration [is] to make each page memorable." That is, the decoration, which changes between and within pages makes each part of the text distinctive and hence easier to memorize. 
4. Thus, according to the evidence from antiquity, it is easier to recall text by remembering the way it appeared in the "original" than by memorizing it, as if it were an isolated string of words devoid of any physical arrangement.
. The quotations from Quintilian are from the Institutio Oratoria. as translated by H. E. Butler in the Loeb Classical Library edition: Harvard University Press (Cambridge, MA and London 1922).
. See especially: Chapter 7 "Memory and the Book," a fascinating account of how illuminations, text, and memory worked together (Carruthers 1990).
Baddeley, Alan. (1990) Human Memory. Theory and Practice Allyn and Bacon: (Boston, etc.)
Carruthers, Mary. (1990) The Book of Memory Cambridge University Press: (Cambridge).
Hartley, James. (1992) The Visual Chunking of Text. PSYCOLOQUY 3(66) reading.11
Neisser, Ulric. (1988) Time Present and Time Past in M. M. Gruneberg, P. E. Morris, and R. N. Sykes, editors, Practical Aspects of Memory John Wiley & Sons: (Chichester, etc.) 545-560.
Small, J. P. (1992) Historical Development of Writing and Reading. PSYCOLOQUY 3(61) reading.10