Retrieval Practice

HOMEWORK is a new strategy

With the changes in curricula across the world, characterised by increased content demands, and many courses moving to exam based summative testing, the retention of knowledge by students has become paramount to achievement. Teachers need to ensure 2 main things:

1. Content taught is secure

2. Content taught can be recalled easily when required

This has enormous implications for homework design.

The underlining premise of retrieval practice is that information that a student has stored in their long term memory needs to be recalled several times before the knowledge could be deemed to be completely secure, and retrievable at later stages. By actively 'working' on the memory, the memory is strengthened.

As retrievalpractice.org state, struggling to learn – through the act of “practising” what you know and recalling information – is much more effective than re-reading, taking notes, or listening to lectures, and so the opportunities that homework present to aid this process are significant.

When information is presented to students it is processed in the working memory.
If it is understood, it moves into the long term memory (LTM).
However, if we don't do anything with it when it is there, it potentially gets lost. What we need to do as teachers is to 'work' on the knowledge in the LTM, by retrieving it,
so that it becomes semantic memory, memory comprised of schemas that we can recall easily.
Once that is achieved, when new information is presented, students have the previous knowledge 'at their fingertips'.

SECURE CONTENT - To assist in the consolidation of learning within the classroom, teachers can use homework as a means of strengthening the knowledge students have been taught. Often, these tasks will be in the form of practice, with questions or exercises designed to master skills. Such tasks should be 20-30 minutes in duration, and precisely matched to the content just taught. Redding (2000) states that homework should not be given on topics that have not been taught, and Cathy Vatterott (2010) is emphatic when suggesting that teachers should not assign homework as a matter of routine, rather, only when there is a specific purpose (cited by Carr 2013). Instructions for the tasks must be precise and clear. There should be an example of a question or exercise available as a model for students to get the ball rolling. These types of tasks should not require students to receive any outside or external help.

Rather than only consolidating learning, there will be occasions when extension or expansion of the knowledge will be suitable. This is covered in the Independent homework section here.

20 - 30 min tasks
Based on what was just taught
Precise instructions
Modelled example to begin

TIPS FOR SECURING KNOWLEDGE

Just because a student has learnt something in a lesson doesn't mean we should be content with that. If a student can't recall the knowledge at a later date, then they effectively haven't learnt it. This is incredibly pertinent if the knowledge learnt is a requirement for the next learning sequence, as most school learning is. Retrieval practice greatly assists in reducing cognitive load: as the learner becomes increasingly familiar with the material, the cognitive characteristics associated with the material are altered so that it can be handled more efficiently by working memory (Sweller 1988).

Homework presents a wonderful opportunity to develop retrieval. The very nature of time in-between the lesson and the subsequent engagement with the material satisfies one of the important characteristics of the concept: the element of forgetting the information. The research by Bjork is clear that the memory is strengthened if the student is challenged to remember it, a likely process if distractions occur between engaging with the content.

Tasks should be 20-30 minutes in duration, test like in nature, and should be a mixture of previous learnt material. Ideally, the majority of the tasks would focus on knowledge learnt very recently, with some percentage of exercises based on knowledge from a few lessons ago, with a final addition of a task based on something from much earlier in the year (or even from previous years). This exploits the concept of interleaving, which suggests that by chunking revision the likelihood of remembering the knowledge increases: again the 20-25 min task works well here, perhaps 20 mins on English, then 20 mins on Maths.

20 - 30 min tasks
Tasks aim to essentially test what is in the memory
5:3:2 ratio (5 exercises form recent, 3 from a few lessons ago, 2 from a long time ago)
Repeat past tasks, but adjust tasks slightly to ensure mastery
Tasks can be quizzes, writing prompts

TIPS FOR successful Retrieval practice

Sweller, J., Cognitive load during problem solving: Effects on learning, Cognitive Science, 12, 257-285 (1988).

Intro to research by Bjork

Professor Bjork - how memory shapes learning

Professor Sweller - cognitive load theory and its implications

Professor Bjork - learning to learn

Carr 2013

Bjork learning and forgetting lab