(don’t) Forget About It: Constructing Knowledge with Retrieval Practice

I’ve been teaching Macbeth the past month and naturally I anticipated that students would struggle with not just the language, but the plot. In addition to the multifaceted characters, there are multiple pieces of information to juggle at all times: What was the historical context of witches? What did the prophecy say? What was X character’s response and how have they changed? (Just to name a few).

So, when \ I saw Jon Gustafson post about retrieval practice earlier this week I realized that 1) I wasn’t being as intentional with retrieval as I needed to be but 2) I was/am using retrieval in other ways that are still of benefit. This post will outline a few ways to meaningfully integrate retrieval practice in the English context and what I changed after seeing Jon’s work.

Activating Existing Background Knowledge

In the reading methods course for my teaching license, we were taught a number of comprehension strategies designed to “activate students background knowledge” without ever being taught what exactly that meant. Really, my instruction amounted to being able to tell students:  “Just activate your background knowledge! Activate it!”. 

Thanks to cognitive science however, we have a much clearer idea of how this actually works. Dan Willingham wrote that we learn things in the context of what we already know. Greg Ashman further explained this as a web of interrelated web of ideas. As Ashman writes, “we need to build schemas carefully and systematically so that students can learn important ideas regardless of their everyday experience”.  Rather than confine our students to a life where they only know what is representative of their immediate surroundings, we should use their existing knowledge to build from and create that web of interrelated ideas. When students learn new concepts in this manner, they strengthen the existing connections, after which more abstract connections and decontextualization may be achieved. 

Constructing the Web of Knowledge With Retrieval Practice

In Cain et al’s study they found that even if all students have the same level of knowledge, less skilled readers still had more trouble generating knowledge-based inferences.  In other words, poor comprehenders were constructing incomplete representations of texts because, as Cain concludes, students were not successfully integrating the correct pieces of information to make inferences.   

We shouldn’t assume that just because we read to students, or because they have read it themselves, that the knowledge floating around will automatically integrate itself and fall into place for analysis. As English teachers, we want students to be able to do the “heavy lifting”, but often forget how much knowledge construction and integration actually has to take place before that can happen. For that deeper learning to take place, we need to help students make “organised knowledge structures”.  Efrat Furst writes that the point of retrieval is just that:  to test the existing loops in different contexts and identify points where the associations are still dependent on the original learning context”. In other words, figure out where the gaps in comprehension are and help students organize the knowledge into place.  We have to explicitly teach students how to construct meaning, and without active retrieval, we are leaving students to unproductively struggle.

Construct, then Infer: Building the Schema from the Ground Up

So taking in all of the research, I incorporate this in my class in a couple of ways:

  1. Daily or weekly discussion of an “opening thought”  and priming the unit with a set of anticipatory statements

What this looks like: Take a situation/theme/concept from the story you’re reading that day, pose it a statement or question, and have students discuss it simply based on their experience. This has students begin to think about what they will be reading or analyzing for the day or unit. When you then transition to discuss the story, you’ll find students making the connection and saying “hey, that’s like what we did in the opening thoughts”. 

This approach  has dual benefits: it helps students construct meaning in a story and take a seemingly irrelevant story, such as many would claim with Shakespeare, Dickens, and other classic literature, and makes it relevant. 

Example anticipatory questions for Macbeth

2. Active Retrieval of Key Knowledge 

In Act 3 scene 3 of Macbeth, Banquo is murdered but Fleance, his son, escapes.  A typical question would be to ask students“why is this scene important”. Simple enough right? Nope. 

The answer we get back is often very surface level like the following :

because Banquo died.

Clearly, this response is merely restating what happened and not analyzing the importance. 

Instead of simply asking the question and having students give back blank stares,  I backtracked, and thought of all of the relevant information students need to have in order to really analyze the scene.:

  1. That the prophecy said though Banquo will not be king, his sons will (1.3)
  2. That Macbeth had planned to “stop” the prophecy by murdering both of them (3.1)
  3. That Macbeth was becoming more mentally unraveled as he commits more murders, but is still seeing murder as his only option (1.5, 2.1, 3.2)

What this looks like: Prior to posing the analysis question, I had students answer the following questions to actively retrieve key knowledge and construct the knowledge structure (we discussed responses as a class afterward):

What did the prophecy say?

What is Macbeth’s plan to “stop” the prophecy?

What happened at the end of the scene?

How does what happened in the scene interfere with the plan?

NOW INFER, based on what we know about Macbeth’s character at this point, why is this an important scene?

The answers are numerous and allowed students to make inferences from multiple points of the story:

Because now Macbeth’s plan is messed up and he’ll have to hire more people to try to kill Fleance

Now Macbeth might become paranoid and kill more people

Now Macbeth might start to unravel even more and become unstable


After I went through this retrieval exercise, a student said, “Dang Ms. Lane… That’s a lot. 

Well yes, it is. But now everyone can feel success with the tasks, not just those who “already” know. 

I’m on twitter @MsJasmineMN

Sources/Further Reading

Cain Study: Comprehension skill, inference-making ability, and their relation to knowledge

Greg Ashman The Implication of Schema

Efrat Furst : Meaning First

Harry Fletcher Wood: Deep learning (2): structuring and organising knowledge – responsive teaching update

Other Stuff

Jon Gustafson’s Blog

Othello Anticipatory Set of questions

Macbeth Anticipatory set of questions

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s