Since developing my understanding of formative assessment and retrieval practice, I have always attempted to ensure that my lessons begin with a clear recap of prior learning (retrieval), coupled with questions about forthcoming learning which allow me to identify gaps in knowledge to support my delivery. It makes sense to kill two birds with one stone, but alas, I may have been doing this wrong for some time.
Before I continue, it is at this juncture that I feel the need to distinguish between two distinctly separate processes that I will be talking about in this post:
‘Encoding is the process of moving information into your long term memory (LTM) via your working memory (WM) i.e. learning new things
Retrieval is the process of pulling information out of your long term memory and into your working memory. This strengthens the memory in the long term.’
(Definitions courtesy of Adam Boxer)
Last week educational psychologist Daniel Willingham shared a research paper (here) arguing that frequent switching between retrieval practice and encoding impairs new learning – even Willingham himself appeared shocked in his tweet.
It is clear from the paper that whilst retrieval practice strengthens learning, it is reported that ‘interpolated’ testing can sometimes impair new learning. In essence, having a starter quiz with mixed learning (prior and new) may not actually be as useful as one might have thought.
I have been able to make some sense of this thanks to a blog by Adam Boxer (here), who reminds me of some of the key differences between encoding and retrieval:
Now, whilst a quiz at the start of a session isn’t really going to support encoding during the asking of questions, it will create extra cognitive load when the teacher clarifies the responses. Moving between retrieval of old information and encoding of new creates an undesirable extraneous load, due to the effort required to switch between the two distinctly separate processes (encoding and retrieval). Furthermore, it is suggested by Davis et al (2017) that mixing retrieval practice with encoding might bias learners’ attention towards relearning the old information, thus impeding new information being learnt.
As a result of this information, it is suggested that retrieval practice (i.e. recapping prior knowledge) be done separately to the delivery of new learning. Your learners may benefit from a separate recap quiz and initial assessment, or by omitting the initial assessment altogether and just concentrating on retrieval prior to the delivery of new information… I’ll certainly be revisiting my practice.