Working Memory: An Experiment

On this page:

Dr. Shona shares an experiment that will help you and your learners understand and experience what happens when our working memories get full.

Next up is an experiment of sorts. Don’t expect it to make sense at first. The purpose is to help you and learners understand and experience what happens when our working memories get full. Because a lot of us think there’s something wrong with us because we can’t remember stuff. Ain’t true at all. So let me overwhelm you on purpose. Just go with it.

You can watch the video here.

Or you can skip all that and just read the transcript below.

Experiment Instructions

  1. Memorize the code that can be found at this point in the video.
  2. Take 30–60 seconds.
  3. Try to replicate the code without looking at it.
  4. Check your results. How many did you get right?
  5. Hide the code again. Now write three-letter words using the code.


How’d you do? Did you even try? When Dr. Carroll asked us to do this for the first time in Brain Week for Abydos International, I quietly refused. I’m usually a rule follower. But I knew there was no point in trying. So I refused. Turns out, that’s a healthy response to an overwhelming and ridiculous task.

Most people can remember about 5–7 of the codes for the letters. A few can remember all 9. George Miller found in his research that folks can remember plus or minus 7 things at a time. Even better if the numbers are chunked in groups of three and four. That’s why our phone numbers are the way they are.

But things get a bit more complicated when we ask folks to spell some words with the code. Folks still have to think of three-letter words that are possible with the letters available. Then they have to remember how to spell the words. And finally, they have to recall the code. Then, they have to do it all backwards to check to see if they are right.

When asked to spell words, the working memory was already full or overwhelmed. Even worse, the working memory could have forgotten all of the code: the working memory dumped the code to use the spaces to think of the words. Since the code wasn’t stored, there’s not much chance of success, unless…

What if there was a way to store the code in a schema that took up only one space in the working memory? The mind could then use the schema to remember the code and the rest of the space to spell and check for accuracy.

Experiment Instructions: Part 2

Working Memory: An Experiment

Now try the experiment again. Visualize the code in your mind to recall the shapes.
Think of three-letter words and translate the letters into the shapes.
Reflect: How was your experience different when using the grid as a schema?

Translating the Experience to Student Reading Experiences

When students are reading, the demands of working memory are still in place. The struggles that most experience with memorizing the alphabet cipher we shared in the experiment are similar to comprehension issues.

When students read, the long-term memory, short-term memory, and working memory are all active. The reading and assessment tasks use different parts of memory interchangeably. Let’s take a look at the reader’s experience in light of how the memory works. (Considerations of social emotional learning would be additional elements of concern.)

In the chart below, I posed situations learners face when reading a digital assessment. In the second column, I brainstormed with me, myself, and I about what might be going on with the memory and the student’s experience. It’s ugly, y’all. You can’t make that stuff pretty.

Ugly Chart of Reading Experience and Memory Types

Situation Memory Type

Learners are presented with a passage and some questions in the assessment screen.

Long-Term Memory: If learners have had instruction and practice with the platform and digital reading tasks, their long-term memory will trigger previous experiences and familiarity with the format itself. Familiarity and previous experience can help with stress.

Learners recognize the genre and characteristics based on the title and text features.

Long-Term Memory: Procedural and Content: Learners use skimming and scanning strategies to set purposes for reading based on genre characteristics. Information about genre is stored in the long-term memory from wide reading experiences online, with physical texts, and through instructional activities.

Learners consume the text.

Long-Term Memory: Automatic Tasks: Decoding, Word Meaning (basic, content vocabulary, derivations), Syntax, Organizational Structure, and Background Information: Ideally, students decode automatically. As they read, English syntax and semantics combine to help students take in what the text offers. Students use background information and experiences to connect to the information in the text. When students read about new content, they connect to previous experiences analogously or not at all.

Short-Term Memory: Students begin to collect information about what they are reading, adding to what they already know or have just read. Some students begin to create a visual-spatial memory about where they are in the reading and where certain elements of the text are placed. Students who experience difficulties with other reading components quickly become overwhelmed, skip content, or let their eyes pass over the text without internalizing or remembering much at all.

Working Memory: As students read, they continually monitor to see if the words make sense in the sequence of the grammar. This is called comprehension monitoring. If there is a problem with meaning or pronunciation, students must stop and problem solve. This stops the automaticity of the comprehension process because learners must switch processes and use additional working memory spaces.

Short-Term Memory: As the text features shift, the reader changes purposes for reading, changing their information search that fits the demands of the text.
Working Memory: Ideally, students note when text features shift. If readers do not notice the shifts, comprehension monitoring is flawed. Learners form incorrect theories about content and use flawed approaches to the text.

Learners consume the questions.

The same processes for consuming text are in place for reading the questions. In addition, a further demand on memory requires learners to connect the questions to the elements in the text, texts, or the question types and demands.

For multiple-choice questions, the memory “sees” each segment as a different item: stem + the number of answer choices. Each must be processed separately and then again with the stem and each answer choice, one at a time.

The learner must hold space in the working/short-term memory for the stem + answer choice + text(s).

Learners critique the text in light of the questions.

This is where the advanced elements of analysis and synthesis, and comprehension: inference, generalizations, etc., play out within the components of memory. If flawed approaches to consuming the text have taken place or have used too much of the working memory, success is limited or even impossible. All forms of memory are working in concert.

Learners critique the questions. Learners problem solve based on characteristics of the item type and text evidence characteristics (including complexity).

This is another advanced collaboration for memory tools. As in the previous row, readers apply multiple comprehension skills along with background, logic, and reasoning skills. All forms of memory are working in concert.

Learners use process of elimination, proof and disproof strategies, and preponderance of evidence to select the most plausible answer.

Readers hold the stem and one answer choice at a time to juxtapose with the text evidence (text-based schema), comprehension, and personal schema. Some readers will remember previous lessons and academic terms. All forms of memory are working in concert. Readers with limited working memory struggle with balancing multiple ideas and text components. If readers have experienced comprehension issues, problem-solving approaches will not match the need.


Some readers can use previous questions about the text and decisions they have made to help make decisions about other items. Learners with compromised comprehension or working memory may not have access to this type of thinking as an asset for decision making.

You see, all of this is very complex. It’s not a matter of sitting down, reading something, and answering a bunch of questions. And what I’ve described here says nothing about the social-emotional realities learners experience just by being alive. Let’s add life experiences and the anxiety of a high stakes testing situation. If all of those things were perfect, the task would be very difficult. And we know everything isn’t perfect. But we do know that success is possible.

As I’ve worked with students who struggle in passing reading assessments, they are relieved when they realize why things aren’t working for them. All along, they’ve just thought something was wrong with them because they struggled. When I complete the cipher experiment and explain how their memories aren’t equipped to hold all of that information at once, they smile. Sometimes, they get mad. “Miss, why didn’t someone tell me this before?” I don’t know, kid. Maybe we didn’t know. I sure didn’t.

Once we are all on the same page about memory, it’s easier for the learners to buy into the strategies for online reading, memory techniques, comprehension strategies, and so on. At this point, students are ready to connect what they know about how their brains work to the approaches they are using with reading and composing texts.

Action and Reaction:

People ask: Is this for my learning, or would you do this with kids? It’s a goose and gander kind of thing. At first, the kids think we are playing a game when I show them the cipher and talk about their memories. As we talk deeper about their experiences, they begin to understand why reading is so exhausting for some of them. We can pinpoint if they are having encoding and decoding problems, language and vocabulary issues, or comprehension, specific ELAR TEKS, stamina and distraction, or attitude/social issues. Each problem sends us on a different path for solutions. Try this activity with your kids and see what they tell you about their struggles.

I’m also interested in what you would change about the ugly chart. Where am I wrong? What did I miss?

TEKS Commentary:

Foundational Language Skills: All foundational language skills must be in place where students function at a self-extending level and have moved from learning-to-read to reading-to-learn. This has to happen before the first high stakes assessment in third grade. If they need more time with foundational language skills – it’s important to add the correct tiers of remediation or intervention.

This lesson really doesn’t have specific TEKS connections beyond the foundational skills because the lesson is more about showing kids how their memories work in concert with reading comprehension and assessment tasks. Our purpose here is to show kids that these things are normal:

It’s normal to read something and not remember a thing.
It’s normal to reread. Over and over.
It’s normal to be unable to pronounce some words.
It’s normal to not recognize what some words mean.
It’s normal to read stuff you don’t like.
It’s normal to read stuff you don’t know anything about.
It’s normal to read digital stuff differently than physical stuff.
It’s normal to encounter texts that aren’t like anything you read for yourself or in class.
It’s normal to read a question and to think you know the answer.
It’s normal to read a question and have no clue of what the answer is or where to look.
It’s normal to be overwhelmed when there’s a lot going on.
It’s normal to not want to read any of it.
It’s NOT normal to do nothing about what’s giving you fits.

My grandpa Sandy had this dog that would sit at his feet while he whittled and chewed tobacco after checking the corn. Grandpa whittled and chewed. Not the dog. Anyway, every once in a while, the dog would shift to snap at a fly and then howl as if he’d been stabbed. When I got enough courage to ask and duck the spittle, Sandy told me that there was an old nail sticking up from the porch boards. Dog was too lazy to get up and move. That’s where I learned that it’s not normal to do nothing about what’s giving you fits.

After I share the cipher and the story of Dog with the kids, they’re ready to hear what I have to say about memory, reading, and how we can make it stop giving us fits.

Next Up:

Working Memory and Overview of Assessment Screen

Please share this with your colleagues:

About the Author

Dr Shona Rose

Shona Rose

Dr. Shona Rose, passionate about literacy and improving student experiences, researches and presents solutions to cause displays of learner growth. These displays become tools for teachers to provide support and intervention to accelerate the impact on student performance.
Dr. Rose uses her experiences as a baker at Kind House Ukraine Bakery, gardening and music, and budding interest as an outdoorswoman and overlander to make concrete connections to literary processes. Her rescue mutt, Joy, and ugly Cornish Rex cat, Youglie, often appear in her writings and activities.
When not researching and reading, Dr. Rose revels in being a “Nona” to her three grandchildren. 
Connect with Shona: 
Or by email to shonarose67 at gmail