Test-enhanced learning: On my terms (Part I)
Test-enhanced learning: On my terms (Part I)

Test-enhanced learning: On my terms (Part I)

For the past three years, I have been reading, researching and thinking specifically about the process of learning and test-enhanced learning within the parameters of Atkinson and Shiffrin (1968) multi-store model of memory, which assumes there are three logically and temporally separate memory stores, and that information is transferred between these stores in a linear sequence. That learning involves “the relatively permanent changes in behavior or knowledge that support long-term retention and transfer,” Soderstrom and Bjork, (2015: 193).

We know that learners, and many teachers for that matter, have a tendency to conflate short-term, in class performance, with long-term retention when, in fact, there is “overwhelming evidence that learning and performance are dissociable,” Soderstrom and Bjork, (2015: 182). What is more, we know that learning can occur in the absence of any performance gains and, conversely, substantial changes in performance too often fail to translate relatively permanent learning. Second, that a multi-store model of memory commonly adopts the notion that large amounts of information are stored semi-permanently in the long-term memory, stored in ‘schemas,’ which provide a system for organising and storing knowledge.

As for learning-remembering, Bjork regularly summarises, ‘using our memory shapes our memories: the things we recall, access, produce or reproduce become more accessible, durable and recallable, whereas ‘the things in competition with the retrieved information become less accessible,’ (Bjork, 2011: 4).

That is a fair amount to contend with for an opening section. To summarise: a multi-store model of memory, learning and performance distinctly different (dissociable), information stored semi-permanently in the long-term memory, stored in ‘schemas,’ and that memory is dynamic.

With “Test-enhanced learning: A practical guide…” almost in the can, the holidays offered an opportunity to wrestle with the two connected, yet separate, conceptual bugbears and the conflation of information and knowledge.

Very quickly, let’s attend to that conflation of information (new content) and knowledge. From this point forward we will refer to new ‘information‘ and information stored semi-permanently in long-term memory as ‘knowledge.’

Teaching your dog to whistle

Next – let’s tackle the first bugbear. The description of memories as ‘information stored semi-permanently,’ is both misleading and somewhat oxymoronic. Second, a challenge to the mistreatment of ‘learning’ as a unitary phenomenon or event. Of learning as a product. That teachers teach and pupils learn. Of teaching as one-and-done. Cue the “I taught my dog to whistle…” gag.

Though worthy of separate interrogation, these two observations above are comorbid and almost impossible to treat independently. Hence it took nearly two months, frequent revisions and various exploratory conversations with Sarah Cottingham @overpractised, to write this series of posts. Here I am on a Saturday evening, making final revisions ready to publish the series, with the book nearing release. If this post or series captures your interest – you may want to scan the book.

Stored semi-permanently? ‘Maintained’ may be more appropriate?

Attention is the bottleneck of learning. Nothing sticks (information) without first being attended to. What is attended to – depends on what is perceived. What is perceived, is heavily influenced by prior knowledge. As a direct result, it is rare to find pure examples of completely disconnected, discrete learning occurrences – although we will give it a go later in Parts II and III.

In my irregular, now-and-again exploratory discussions with Sarah Cottingham, this has been our unvalidated conclusion. To which we have more recently added “learner motives” or “what learners value” also contributes to what information is attended to (See the series Motivated by motivation and Expectancy-value theory of achievement motivation (Wigfield & Eccles, 2000)). Each influencing learner attention and commitment to learning. Without directed and conscious (motived/valued) attention and the regulation of mind wandering, learning is significantly inhibited.

Side note: Whilst you are reading this post, I am acutely aware that I am in competition for your attention with many other opportunities – Thank you for joining and staying with me.

With your committed attention, let’s attempt to encode some new information and then move some information into long-term memory “storage?” as knowledge. Before we do that, let’s pause for a moment to consider the purpose of memory in the first place.

Memory is probably not for us to reminisce about past events. It’s primarily for us to use to make accurate predictions about what might be about to happen. We don’t go around passively taking in our environment – that’s a recipe for getting caught out. Instead we use what we know to make sense of what we experience. In other words, in order to encode what we experience, we are constantly retrieving relevant things we know from memory. Encoding involves retrieval.

Sarah Cottingham

It is important to recognise that any form of perception, attention, that leads to encoding/learning of new information, starts with a combined retrieval-encoding of this information with prior-knowledge. Even where that retrieval is unsuccessful. (What is a little too complex to fully explain here is that even that unsuccessful retrieval has the potential to benefit subsequent learning and long-term retention).

My point here is, rather than simply retrieving knowledge stored semi-permanently in the long-term memory, knowledge is accessed, encoded and retrieved almost simultaneously. As a teacher you will know “that look,” the tilted head, the I-am-working-it-out, trying-to-connect-the-dots, expression. The very act of attempting to retrieve knowledge, successful or not, modifies that knowledge. And that feels far from ‘sedentary.’ More organic and less unitary. Memory is, as I make the case for in the book, ‘reconstructive rather than objective.’ A process and not an event. As Sarah Cottingham presents our memories are:

More ecosystem than a library system.

Sarah Cottingham

Knowledge then, is accessed, (modified and) maintained and reconstructed rather than retrieved and neatly returned into storage. Dr Steven Pan, an expert in cognitive and learning sciences uses the terms “reactivates and strengthens.”

We will explore that ‘ecosystem’ in Parts II and III.

On my terms (Part I)Information is not knowledge (Part II)Knowledge is not information (Part III)Successive Relearning (Part IV)


  • Memory serves futures predictions
  • Attention is the bottleneck of learning. Prior knowledge informs perception, and perception directs attention
  • Encoding/learning, starts with retrieval-encoding based on what is attended to (even without prior knowledge)
  • Information and knowledge are distinctly different
  • Memory, and therefore learning, is ‘maintained’ rather than stored semi-permanently, ‘reconstructive rather than objective’
  • Learning is a process and not an event
Atkinson, R.C. & Shiffrin, R.M. (1968). Human memory: a proposed system and its control processes. In K.W. Spence & J.T. Spence (Eds.), The psychology of learning and motivation: Advances in research and theory. (Vol. 2). New York: Academic Press. Pp. 89–195.
Soderstrom, N. C. and Bjork, R. A. (2015) Learning versus performance: an integrative review. Perspectives on Psychological Science, 10(2): 176–199. https://doi.org/10.1177/1745691615569000
Wigfield, A. and Eccles, J. S. (2000) Expectancy-value theory of achievement motivation. Contemporary Educational Psychology, 25(1): 68–81. https://doi.org/10.1006/ceps.1999.1015


  1. Pingback: Test-enhanced learning (Part II) – Edventures

  2. Pingback: Test-enhanced learning (Part II) – Edventures

  3. Pingback: Test-enhanced learning: Information is not knowledge (Part II) – Edventures

  4. Pingback: Test-enhanced learning: knowledge is not information (part III) – Edventures

  5. Pingback: Test-enhanced learning: Successive Relearning (part IV) – Edventures

  6. Pingback: Reconstructing memory – Evidence for Educators

  7. Pingback: Memory – what comes before knowledge? – Edventures

Leave a Reply