Scrolls and Textmapping -- for Strategies and Content Instruction

From time to time I post about my work on comprehension and strategies instruction.  Here is something that might pique your interest:

Uploaded image from Dave Middlebrook

The image is cropped here.  You can see the full image here:

The photo shows a white board at the end of a college-level Educational Psychology class.

In this example, I walked the students through the process of surveying the chapter (SQ3R) in preparation for reading the chapter on their own.  At the bottom of the white board in the photo, you will see a paper scroll of the chapter.  Above the scroll is a map.  The map tracks the scroll, and shows the breakdown of sections and topics.  Everything is much clearer to students when presented this way.  We get more done -- and we get to go much deeper.

As is the case with any snapshot taken at the end of an hour-long class, the shapshot fails to capture the sometimes wandering and convoluted process that led to the end-view that you see.  During the class, both the chapter structure and content were discussed in great detail, as were a number of strategies for comprehending the content.  As the discussion progressed, the map evolved -- and often changed.  The students were prompted to discover the chapter's structure as well as the various pathways through the text -- to...

  • ask (and find the answers to) questions

  • make (and test) predictions

  • draw (and challenge and revise) inferences

  • connect information across the chapter as well as to outside readings

  • determine importance

  • summarize and synthesize.

They did all this BEFORE reading the chapter -- which is to say that we spent the entire class prereading (the Surveying and Questioning parts of SQ3R) the chapter.  It was an intensive process.  When they left class, they were ready to read the chapter through on their own.

The goal of this kind of instruction is to get students to the point where they can do this kind of intensive pre-reading (thinking, strategizing, and planning) without in-class help.  It is a long process, and really has to happen across the curriculum, in all subject areas in which textbooks and tradebooks are important sources of information.  One of my inspirations for this work is Mortimer Adler's classic, "How to Read a Book".  It's old and cranky (by internet standards), but a must-read.

Scrolls and textmapping comprise a way of presenting and working with text content that is richly multisensory and absolutely explicit -- and by "explicit", I am referring to the original meaning of the word, which is, quite literally, "unrolled", "fully-revealed", and "wide-open to understanding" (from ex + plicare).  You simply cannot get more explicit than an unrolled scroll!

Added to this is textmapping: The process of mapping the text is easy for students to follow.  It is active and very engaging.  Most important, it strongly reinforces the instructor's efforts to share his/her thoughts and thinking process (think-alouds).  The net effect -- and this is important -- is that strategies for learning replace content as the centerpiece.  This how it should be.

This is a kind of instruction that many college students desperately need (and that middle and high school students should be getting, in preparation for college).

No other book form -- ancient or modern, print or digital -- can match the unrolled scroll's powerful instructional benefits.  Scrolls are uniquely suited for classroom instruction.

More information:

I hope that you will try this in your classrooms, share it with your colleagues, and post your comments or questions here (if you have any).




Dave Middlebrook
The Textmapping Project
A resource for teachers improving reading comprehension skills instruction.   |   Please share this site with your colleagues!
USA: (609) 771-1781
linked in:



Thank you for your post. Seems to me that the information was part of a previous long discussion, perhaps a couple years ago, on a discussion board. (But then, my memory always surprises me at the errors I find relying only on memory.)

Since graphic organizers are so important to improving learning and comprehension, alternative methods are important for us to develop and test. I appreciated Dexter and Hughes' meta-analysis of effects of graphic organizers and students with learning disabilities (Learning Disability Quarterly, 2011, 34 (1) 51-72. Certainly some types of graphic organizers (e.g, cognitive mapping, semantic mapping, semantic feature analysis, visual displays, and content enhancement routines) have a stronger research base than others.

A question for us is the extent to which graphic organizers  are integrated into instructional practices for English/reading, science, social studies and mathematics and do the applications yield similar effects as the research developers achieved?


Here's an interesting research opportunity in the area of graphic organizers:

All of the other graphic organizers that have names ending in "mapping" are actually diagramming methods.  That's fine.  They do what they do, and the research has found that they do it well.  But there is a substantive difference between maps and diagrams.  Maps focus on size, dimensions, and spatial configuration -- i.e., on how chunks of gaphic information are assembled in space.  Diagrams focus on connections and sequences. Without exception, all of the mapping-named graphic organizers that have been studied over the past 40-or-so years have been diagramming methods.  There haven't been any true mapping methods -- until now.  For researchers who study graphic organizers, this should be a very enticing development.

The kinds of maps of text content that I am talking about -- true maps, not diagrams -- represent content in a way that has never before been done.  You will not find research on this.  It is new ground.  Textmaps open new opportunities for some very different ways of working with, making sense of, and thinking about the information in books.  You can begin to see this if you study the single photo that I posted earlier (See the cropped photo above, or see it uncropped at  In addition, you can learn more on this introductory page, as well as by reading some of the unsolicited comments that I have received from educators and students.

My work with scrolls and textmapping takes graphic organizers into new territory with lots of potential for new insights and understandings about reading and the ways that we teach both strategies and content in our classrooms.  I would be very pleased if you were to take an interest in exploring this potential a bit further.  I can promise you that it will be well worth your time.  I'm not looking for commitments.  Let's just get started with a conversation, and see what develops.

Interested?  I look forward to your reply.  Either here, or by email.

Dave Middlebrook


Hi Virginia.

Great example of a diagramming method!  It's all nodes (rendered as boxes) and connections.  The nodes are blank (for now) because this is a planning framework -- a diagrammatic framework -- for an as-yet unwritten essay.  The dimensions of each node are the same, as are the distances between each node.  It's purely a conceptual model -- a diagram.  There's nothing (yet) to map.  The essay doesn't yet exist.

Once the essay is written, the essay as a whole will assume a certain form and set of dimensions -- as will each of its parts (main idea #1 plus supporting details, main idea #2 plus supporting details, main idea #3 plus supporting details).  The nodes in the diagram will become sections on paper (or on the screen).  You might be able to see this in the photo (above) of my map of the textbook chapter.  Once the essay is written, each of the sections might well look very different.  For example, one section (corresponding to a node in the diagram) might contain illustrations while another might not; one section might have long paragraphs while another might have short paragraphs; the supporting details in one section might be more complex, and so might be divided into more sub-sections than the other sections.  In other words, the arrangement of each section's interior objects (sub-headings, paragraphs, illustrations, and the like) will be unique.  As a consequence, the sections will not all be the same length; their dimensions will be different and the distances between them will all be different; the visual-gestalt will be different  The essay "map" shows none of this -- and we can't expect it to, because it is a diagrammatic framework for planning an essay.  It's not a map.

The essay "map" (really, it's a diagram), is a great tool for planning an essay.  But when it comes time to really dig in and edit and rework our essay -- or to comprehend someone else's work -- the diagram is a less-than useful tool.  That's where the map really shines.

Look at the essay "map" (diagram).  Compare it to the photo of my textmap (a true map) of the textbook chapter (above).  Notice how my map represents the chapter in spatial terms -- how it sees the main points as distinct areas, each having its own unique size, dimensions, and internal spatial configuration.  Notice how different the representation provided by the essay "map" (diagram) is.

Here's a resource from my site on this topic:

Does this help?

- Dave Middlebrook  


Hello Dave,

Cartographers have created and then, as the mapping craft developed, used and adapted visual conventions for mapping geography, for example, symbols or other representations for rivers and other bodies of water, mountains, political boundaries, and state capitols. I wonder if there are text mapping conventions and if these are taught to new readers as part of the text mapping process. If so, do you have a set of these conventions you could point me to or send as a document?


David . Rosen

Hi David,

Interesting question!  Lots to say on this.  Here are just a few items, to get you started.  These conventions, such as they are, are idiosyncratic -- and hence, not really conventions.  But they are a start.  In time, I hope to see them develop into something more...conventional.  So here they are:

1. Colors: I use brighter/bolder colors for big ideas and softer/quieter colors for smaller ones. I like... 

  • red for delineating the highest-level structural divisions -- such as beginning | middle | end; front matter and back matter; intro | main body | conclusion
  • green for sections and their corresponding headings (if any)
  • blue for sub-sections and their corresponding sub-headings (if any)
  • purple for sub-sub-level
  • yellow for key words, phrases, sentences
  • gray or brown for illustrations, charts, maps, tables, side bars, etc., along with captions
  • for more complex texts, I add colors for more levels and/of kinds of details

2. Lines

  • line thickness: (sometimes, but not always) Thicker lines for bigger ideas; thinner lines for smaller ideas and details
  • I use a boxes-inside-boxes model -- i.e., sub-sections fit inside sections
  • sometimes I go back and thicken the bounding lines that surround important sections
  • I only highlight the yellow stuff (see colors, above).  For everything else -- such as headings -- I draw a box or circle around each instance.
  • I draw arrows to connect stuff.  Sometimes the arrows that I draw are well-mannered and respectful -- i.e., I route them around the intervening text.  Other times I simply take the most direct route -- i.e., I draw a straight right across the intervening text.  It depends on what the arrow is indicating, and what's going on in the intervening text.

3. Symbols: These are, as you might expect, highly context-sensitive, so I usually make them up at the time, for a particular text.  I reuse symbols all the time, but I rarely use them the same way twice.  It's very idiosyncratic.  If I think it's needed, I'll draw a key on the board, to remind everyone (including myself).  

  • If I'm modeling strategies, I place simple symbols in the margins.  Usually circled.  A question mark to indicate a question in the text; a "C" for connections -- and tt for text-to-text connections, ta for text-to-author, ts for text-to-self, etc.; a left-facing arrow for a backwards-looking inference; a right-facing arrow or a "P"for a forward-facing inference (prediction) with an arrow going forward to where the prediction either falls apart or turns out to be on tartget; just a colored-in (typically sloppily scribbled-in) circle for a sensory image; a bold arrow for important (or a bold line -- see "lines", above); an "S" for parts relevant to a particular synthesis
  • If I'm teaching a novel, I might today use "S" for setting; "C" for character; "T" for time; "P" for plot, but tomorrow it could be different.
  • If I'm teaching a history text, I might use checkmarks for key episodes leading up to an important event or development.  Again, it's all very dependent on context -- the content, what must be learned, etc.

So those are my very own, wonderfully idiosyncratic conventions.  I hope this is useful to you.

- Dave Middlebrook 


Thanks, Dave.

Here are some follow-up questions:

1. With my question about conventions I hoped to get some insight into these two related questions, "What do you think readers need to know about the organization, layout and other features of text?" and "What do you think text mappers need to show them (teach them)?"

You have already provided some answers to those questions, but I wonder if there are other features that you text map that you haven't mentioned yet. From what you have written, I gather that you have some separate text mapping strategies for fiction and non-fiction as well as some that may apply to both fiction and non-fiction. Is this correct?


  • Setting
  • Character
  • Plot -- and key episodes of a plot


  • Big ideas and small/smaller ideas

Both fiction and non-fiction:

  • Structural divisions such as beginning, middle and end; intro, body, conclusion. (What do "front matter" and "back matter" mean?)
  • Connections (indicated by arrows) (What kinds of things do you connect?)
  • Questions (Are these the author's explicit questions, implied questions; or are they a reader's questions?)
  • Inferences (Author's? Reader's? Both?)
  • Predictions (Author's? Reader's?  Both?)

Have I accurately categorized the text features that should or could be mapped? If not, how would you change or add to this?

2. I am especially interested in what text mapping would look like for for adult neo-literates who read at a roughly 3rd-4th grade level, who are getting meaning from text, but who read slowly, and who have a limited vocabulary and are not fluent)

3. Has a video been made of the text mapping process? If so, with adult learners? I would be interested in seeing the process in action with learners.


David J. Rosen




Hi David,

In reverse order:

#3. No video yet of adult learners.  Would you be interested in collaborating?

#2. I'll give you an example: Let's say you think some or your students might be interested in an article from National Geographic.  It's way above their reading level, but it offers a lot of visual support (photos, maps, etc.).  And a lot of adults find it interesting.  So you make a scroll (buy 2 copies of the magazine; zip off the spine with a razor knife; arrange the pages; tape them together) and tape it to your blackboard.  And let's say that you want to focus on vocabulary development.  So you pick a few words for them to search for -- names, places, whatever seems appropriate.

  • You ask them to circle or highlight each instance of each word.  Use a different color for each word.

  • Have them take turns using each color -- in effect, checking the work of the previous person and finding any stray words.

  • Then you have them step back and look at the article as a whole -- much as they might look at a painting on the wall at an art gallery or museum.  Have them talk about the word-instances: Where are they?  Are they clumped together or spread out?  What can they infer from this?  What's this article about?  What do the illustrations have to do with the words?

  • Then point to some headings.  Ask them, Are all these headings at the same level?  Do we have any sub-headings?  How do we know?

  • Have them decide on colors to use for each heading level, then have them circle each heading with the correct color.  As they did with the words, ask them to check each other's work.  Ask them to explain why they used a certain color to mark a certain heading.  What visual information can they use to make these decisions?

  • Ask them to read aloud the heading and discuss what it means.  How does this relate to the article as a whole?

  • Then have them draw a bounding line around the text in each section.  Leave the illustrations outside the bounding line -- free-floating, as it were.

  • Now stand back again and look at the article as a whole.  Look for patterns -- certain word-instances in certain sections.  Does this improve their understanding of what to expect from this article?

  • Have them take turns reading aloud some of each word-instance's surrounding text -- perhaps just a few words in either direction, or maybe entire sentences.  Does this improve their understanding of what to expect?

  • Have them share their thinking -- their questions, connections, inferences, and predictions.  For each illustration, ask them to draw a line to the part of the text to which it relates.  Have them share their thinking about these connections.

  • All along, try to move back and forth between visual thinking and textual thinking.  Sound out words selectively.  Only decode enough to keep the conversation developing.

  • Let them use the illustrations and typography to guide their focus.  The goal here is to use the comprehension piece to keep them engaged and to motivate them to decode, bit by bit, more words.

  • The overall experience should be engaged, discovery-based, conversational, and -- despite the decoding piece -- not at all dumbed-down.

#1. So what do readers need to know about organization, layout, and other features?  I assume that you mean this with respect to textmapping, not text structure (the research is clear on the importance of structure and strategies for using it).  So if we're talking about textmapping, the answer to your question is, They need to learn to see the typography -- all of the visual elements on the page, how those elements are arranged, how all the pieces fit together to make a visual whole.  They need to see the fonts and styles and sizing and spacing; they need to see how the paragraphs are laid out and how the illustrations are integrated (or not integrated) into the flow of text that moves down the length of the scroll.  These things are impossible to see across page breaks in a bound book, but very easy to see on an unrolled scroll.

And what do "text mappers" (teachers? and fellow students) need to teach them?  Just model the process.  Model it repeatedly.  Think aloud about your process.  Gradually turn over control of the process to your students, until they finally are doing the entire thing themselves.  Don't worry about sticking to a fixed set of mapping conventions.  Instead, let them map the text as they wish -- in effect, showing a trail of their thinking using lines and symbols and any other markings that they find useful.  As long as they can explain their thinking and demonstrate that they comprehend, that's all that should be required.

And you asked a few smaller questions:

  • other features: I sometimes map review items in textbook chapters.  I number the items and then have the students find the relevant references in the chapter -- such as definitions for key words, answers to literal/factual questions, and bits and chunks of text that are germane to answering questions of inference or synsthesis.  And there are many more.  It really depends on what I find and what I need -- i.e., it depends on the text and on my reading (or teaching) purpose.

  • separate strategies for fiction and non-fiction?  It depends.  And some that work on both?  Yes.  Still, it always depends on the text and my reading or teaching purpose.

  • front and back matter?  The stuff at the beginning and end of a book.  Most books have rather pedestrian front and back matter -- if any at all.  But there are also wonderful examples.  First thing that comes to mind is a dictionary.  Very cool front and back matter.  And children's books -- check out the illustrations inside the covers and on the title and half-title pages.  Sometimes there is some amazing stuff there.  That said, bland front and back matter can be very useful.  It just depends.  You have to look, and you have to think about it.

  • arrows: I connect related stuff.  If I see a connection, I'll draw an arrow if I think it will help me.

  • questions, inferences, predictions -- author's or reader's? explicit or implicit?: All of the above.  And I would recommend distinguishing between them!

OK.  That was a lot.  I hope it's useful.  If you are intrigued -- and I hope that you are -- we could always talk by phone.

- Dave Middlebrook |

Glad you are interested!  Lots of potential for ABE in Ohio.  Here's a hands-on assignment that you can try: Look for "Your Homework" under "Read.Write.Act 2012" on the following web page:

Do the homework and then check back with questions.  The reading level is, I suspect, much higher than what you would normally use, but you'll get a feel for using the scroll.  It's a good way to get your feet wet.

By the way, my Read.Write.Act presentation was cancelled due to Hurricane Sandy.  I'll be rescheduling it for after the holidays.

Thanks for your interest,

- Dave Middlebrook | 

Hello Dave,

Thanks for taking the time to carefully and fully answer my questions and to make the text mapping process clearer.

Here are some thoughts about this process based on my interpretation of your responses.

Many math teachers know that when learners struggle with a math problem in a group both the struggle itself -- the application of collective peer knowledge of math facts and problem solving strategies -- and the testing out of these strategies together can be very helpful to all the learners, often more helpful that a teacher demonstrating how to do the problem.

Some English language teachers who use smartboards (electronic whiteboards with Internet access) have found that adult immigrants who have never touched a computer before find this very public way of watching other learners use a large smartboard keyboard, of navigating a web page in front of the class, and of finding things on the Internet on the whiteboard screen to be a low-risk, highly engaging way to learn digital literacy skills. They find that when their students do begin to use them, computers are much less threatening.

Text mapping, like these other kinds of guided group learning processes, also appears to enable a group of learners to work together to unlock the features of a particular kind of text, from a particular genre, using _all_ the clues in the ext, and all the knowledge and reading experience of the group to do so. It seems to me to be a constructivist (project-based) approach to learning.

Team problem-solving is what people often do at work. There is a problem. A team of workers has to solve it. They draw on the intellegence and experience of the group. If the problem involves getting meaning from text, they figure out the solution together using text tools such as front and back matter table of contents, index, or glossary, and maps, illustrations, charts, diagrams, and others.

I see a lot of potential for using text mapping with adult basic skills learners in generic classes and in workplace basic skills programs. I am eager to see it in action.

David J. Rosen

Hi David,

I'd love to demo this for you.  Perhaps we will be able to arrange that.

Your comments are right on target.  Scrolls and textmapping create a learning environment that supports a constructivist approach -- discovery-based, cooperative, lots of conversation, thinking aloud, and sharing ideas; lots of peer interaction.  This is true whether you're teaching course content, writing/composition, or the elements of reading (such as vocabulary and word knowledge, grammar, punctuation, and comprehension skills, strategies, and routines).  Scrolls are simple and low-tech, but you can do a lot with them in the classroom.

- Dave Middlebrook |

Something to think about: You asked about "explicit" versus "implicit" connections, etc.  We use these words all the time -- "explicit" and "implicit".  But what do they really mean?

These are ancient words.  Explicit comes from Latin -- from ex + plicare.  Literally, it means "unrolled" or "unfolded".  The image was of something -- such as a carpet, map, or scroll -- that was laid out, fully revealed, and thus wide-open to understanding.

"Implicit" is the opposite -- rolled or folded back into itself.  The modern book -- based on the Roman codex -- is literally, physically implicit.  A single sheet of parchment was folded back into itself several times to form a quire.  Once the folds in the quire were cut, you had a stack of pages.  Many quires, bound together, made a codex.

The moral of the story?  Bound books are literally implicit.  They are not laid-out.  They are not fully revealed.  They are anything but wide-open to understanding.  Unrolled scrolls, in contrast, are absolutely explicit -- which is what makes them so uniquely suited for the classroom.

- Dave Middlebrook |