There has been much more understanding around the importance
of retrieval practice to strengthen the durability of knowledge in pupils long
term memory but whilst a lot of books and blogs have discussed how to
ember retrieval practice (be it allow stakes
quiz, self quizzing, flashcards etc) what I have found the most interesting as I’ve
tried to embed this into my practice is considering what to retrieve.
In a sense there is no bad knowledge to retrieve. In a well-designed
curriculum, everything we teach should have significance and we want pupils to
remember all of it but some knowledge is more equal than others when it comes
to retrieval.
Like many teachers (and a la Rosenshine) I start every
lesson with 5 or 6 questions about prior learning that pupils complete with no
notes. Initially, the topics these questions would be based on would be random
but following some trial and error I now consider 3 areas when deciding what
topics to base my retrieval on:
1. Prerequisite Knowledge
If I am going to teach indirect taxes to solve market
failures, I need pupils to have a strong understanding of negative production externalities
and how increases in costs of production impact supply curves. Given this, some
retrieval questions will inevitably be around this.
Basing retrieval on this (with feedback then given) will enable
pupils to access new learning. If this knowledge is more secure in their long-term
memory, pupils can devote their working memory purely to the new information I
am about to teach them. They already understand higher costs leads to an inward
shift of supply and what the market failure caused by negative externalities
are, so now all they will be thinking about is how an indirect tax can fix
this, thus avoiding cognitive overload.
When circulating, if pupils are struggling with this, then
it is an opportunity to re-teach when going through feedback. This is also of
benefit to those pupils who were absent when that topic was covered so that
everyone can access new content. Circulating when this retrieval is going on
will also give you data about which pupils you will need to support more as you
introduce new information and guide pupils towards independent work.
2. Spaced or Distributed Learning
Some of my questions will have nothing to do with what they
are about to learn. Instead, I will consider 2 things, how often have they retrieved
this information and how long has it been since it was last retrieved.
There is nothing new in the idea of ‘forgetting curves’ (Ebbinghaus
discussed it in 1885!) but the simple idea is that for pupils to be able to keep
and then retrieve information in their long term memory they ned to retrieve
information multiple times over a prolonged period of time.
If I taught the concept of cross elasticity of demand 3
months ago, I then ask them to retrieve it again as enough time has passed that
they are likely to start forgetting. Ideally you want retrieval to be spaced to
the point where they retrieve it again just before they forget the information,
so it is of maximum difficulty while still doable.
Of course, we don’t know when they are just about to forget
and can’t be overly scientific about these things so I just consider how long
it has been since this information was last retrieved rather than spending too
much time calculating a formula of exact times since last retrievals.
But it is not just the gaps in between retrieval that needs
to be considered, it is also how often it needs to be retrieved. One of the
first concepts I teach at A Level is opportunity cost. If over a year it has
been retrieved 5 times, is that worth then continuously retrieving again? It
can be included in self quizzing that pupils complete independently but there
reaches a point where knowledge is secure in long term memory.
So, it is both the time since knowledge has been retrieved
and the frequency of that retrieval I consider when setting retrieval questions
to ensure spaced learning but there is one important exception to that rule.
3. Automated Knowledge
There is some knowledge that is so fundamental to pupils understanding
of a wide curriculum that they don’t just need to know it but it needs to
become automatic. At GCSE level, pupils need to understand the concept of
market equilibriums and how shifts in supply and demand impact these.
They are going to have to use this knowledge to explain
changes in the equilibrium for product, labour, capital, and money markets
throughout their 2 years. This means it’s not enough that they can retrieve
information they need to do it automatically as though I am asking them their
name.
What I do is consider for each specification I teach what
are those key threshold concepts that pupils must have an automated knowledge of
and questions around these feature prominently in my retrieval, even if pupils
can answer them confidently.
These are the areas where we don’t practice until
we get it right, we practice until we can’t get it wrong!