Monday, February 27, 2012

Why don't we talk about mental models?

When I first learned the term "mental model", it made perfect sense to me. A mental model is the reader's mental representation of a text--the amazing combination of ideas from the text and the reader's background knowledge. In the classroom, I see it as my task to help readers develop rich mental models of the texts that we share and their own self-selected texts.

Thinking about what kinds of mental models my readers have created--and where their comprehension problems might be--guides my instruction. For example, this week some of my struggling readers completed a cloze passage. Looking at their responses, and thinking about what clues they were or were not using, helped me to figure out where to go next. I could see which students were reading sentence by sentence, and which students had developed a more global view of the text.

Talking about mental models also helps me to understand what students are thinking. We were talking about the "Invasive Species" text using the click or clunk idea. Some students told me that this sentence was a big clunk for them:

Garlic mustard escaped from gardens.

When we talked about it, I found out that they were visualizing the actual garlic mustard plant walking out of the garden! (What a great mental image!) However, they knew that this couldn't happen, so the sentence "clunked"--that is, they didn't understand it. Getting them to talk about this mismatch in their mental models helped me to guide them toward the inference they needed to make in order to understand this sentence--the garlic mustard plants didn't escape, but the seeds did.

But not many people around me are talking about mental models. In fact, in the entire Pennsylvania Comprehensive Literacy Plan--a document meant to guide all Pennsylvania schools in their literacy instruction--the term "mental model" is mentioned only once, as an aside in a bullet point about the values of read aloud.

So why don't we talk about mental models more? Is it because there is no quick and easy test? We can't graph it, we can't count it, so it must not exist? Or are the conversations going on, and I'm just now aware of them? I'd love to know!

Find out about mental models:
Building mental models (blog post)
Help for word callers (blog post)
Overview of situation model
Article about mental models and visual and verbal processes
 Article from Project Read


  1. Excellent info. I'm going back to the links you posted and reading the posts you wrote before and it is really helping me understand how I can help my struggling readers more. I have a question for you. I know you write many of your short reading texts yourself. How do you determine the readability level?

    1. Hi Nikki! I figure out reading level by combining the Okapi Readability Tool (, the Flesch-Kincaid Readability Scale available in Word, and what I know about my readers. It's fascinating to see how scales differ (try typing some of your state test released tasks into Okapi to see what you find!) and how nonfiction always comes out higher than fiction.