Rosenshine and Sherrington on Reviewing

Jonathan Beale, Researcher-in-Residence, CIRL

Reviewing is important in teaching and learning. We forget things, but regularly revisiting what we have learned can reduce the amount we forget. Reviewing also helps us develop our understanding of what we have learned and consolidate knowledge in our long-term memory.

Daily, weekly and monthly review

In his ‘Principles of Instruction’,[1] Barak Rosenshine emphasises the importance of reviewing in effective teaching and learning. The first and final of his ten principles concern reviewing:

1. Daily review

‘Begin a lesson with a short review of previous learning: Daily review can strengthen previous learning and can lead to fluent recall’ (Rosenshine, 13).

10. Weekly and monthly review

‘Engage students in weekly and monthly review: Students need to be involved in extensive practice in order to develop well-connected and automatic knowledge’ (Rosenshine, 19).

Automatic recall

In the above principles, Rosenshine refers to the stage we should aim at when reviewing: to be able to recall prior learning fluently and automatically. Automaticity denotes the stage in the learning process where learning and practice has been undertaken such that recall is effortless. Rosenshine’s phrase ‘fluent recall’ also relates to automaticity. Fluency concerns our ability to be able to quickly and easily recall prior learning, without having to make a significant effort to remember – for example, without the need of prompts such as notes. The highest form of fluency is where recall is automatic. The aim of reviewing is to reach the stage where the learner can recall their prior learning in the area being reviewed with little or no effort.

As we saw in an earlier blog post, to achieve automaticity, we need to ‘overlearn’: to learn or practise beyond the point of ‘initial mastery’, such that recalling existing knowledge of the area in question is ‘automatic’ and our skills are fluent (Rosenshine, 13 & 18). Rosenshine argues that overlearning is required for any skill to become automatic (Rosenshine, 18).

The key benefit of reaching the level of automaticity with recall is that when this stage has been reached, engaging in the task or skill that has become automatic takes up less of our working memory – the area in which we process information. Rosenshine writes, ‘When material is overlearned, it can be recalled automatically and doesn’t take up any space in working memory’ (Rosenshine, 18). You thereby put yourself in a better position to learn new things or problem-solve within the area where you have made recall in that area automatic and your skills fluent.

Sherrington’s third ‘strand’

In his book Rosenshine’s Principles in Action,[2] author and educator Tom Sherrington divides Rosenshine’s ten principles into four ‘strands’. He uses the strands to explain Rosenshine’s principles by connecting the principles with those to which they bear the closest relations, to illustrate how the principles complement and support one another. Sherrington’s third strand, ‘Reviewing material’ consists of the above two principles.

In his ‘Principles’, Rosenshine lists seventeen ‘instructional procedures’ which he then condenses into his ten principles. Sherrington’s third strand concerns the following of Rosenshine’s instructional procedures:

  • Begin a lesson with a short review of previous learning.
  • Re-teach material when necessary (Rosenshine 19; Sherrington 35).

Sherrington outlines the following general advantages of reviewing. First, reviewing reduces the amount students will forget. Second, reviewing supports retrieval practice; and ‘Retrieval practice supports building our long-term memory and our level of fluency in recall’ (Sherrington, 35). Third, learning is a generative process: when we review prior learning, we generate new ways of understanding what we have learned – for example, by connecting it with new areas of learning, or with concepts or subject matter we haven’t connected it with before, or by relating it to different experiences or memories, etc. Sherrington writes:

‘Daily, weekly and monthly review activities give students opportunities to generate versions of what they know and understand, helping to strengthen future retrieval of the knowledge involved, build fluency, and identify where they might have residual gaps or areas of uncertainty.’ (Sherrington, 39.)

Sherrington’s guidelines for effective reviewing

Sherrington recommends that reviewing should be part of a regular teaching routine. Drawing upon Rosenshine’s ‘Principles’, Sherrington outlines the following seven guidelines for making daily, weekly and monthly reviewing ‘part of an effective and sustainable routine’ (Sherrington, 40)

1. Involve all students

Good reviewing techniques should ‘involve all students checking their knowledge’ (Sherrington, 40). Contrast this with questioning, where individual students might be asked about their learning in a particular lesson, and a different set of students in another lesson.

2. Make checking accurate and easy

It should be possible for students ‘to find out what they got right and wrong, what they know well and where they have gaps’ (Sherrington, 40). Good reviewing techniques ‘involve students testing their knowledge and then checking … for accuracy and completeness’. Sherrington notes that this applies more readily to simple retrieval activities; it doesn’t apply to more complex tasks, such as ‘giving students extended mark schemes to mark longer assessments’ (Sherrington, 40).  

3. Specify the knowledge that retrieval should be based on

Students should ‘know the set of knowledge any retrieval will be based on, so they can study, prepare and self-check’. Sherrington adds that it ‘must be possible for students to check their own answers’ (Sherrington, 40). As a process of self-assessment, reviewing should be conducted by students independently to a significant degree.

4. Keep learning generative

Sherrington suggests that students should ‘think for themselves’ and ‘explore their memory to check what they know and understand’. Students should be expected to recall prior learning without prompts. So, he recommends employing retrieval practices where there are no memory prompts, avoiding resources such as cue cards, scaffolds, cheat-sheets or open books.

5. Vary the diet of teaching and learning methods

Employing a variety of teaching and learning methods helps ‘students to explore their schemata in different ways, strengthening future recall’ (Sherrington, 40). A ‘schema’ isa well-connected network of ideas (Sherrington, 19); reviewing should provide students with the opportunity to explore and consolidate their ideas and prior learning in a variety of ways.

6. Make retrieval practice time efficient

The final two guidelines Sherrington offers concern time and workload. His sixth guideline is don’t let retrieval practice dominate entire lessons.

7. Make retrieval practice workload efficient

Sherrington’s seventh guideline is that we should avoid allowing reviewing practices to create unsustainable teaching workloads. Sherrington writes that ‘The best methods do not involve the teacher checking the students’ answers. … A teacher might choose to check the occasional test but for routine practice, students should do it themselves’ (Sherrington, 40). So, the most effective reviewing, according to Rosenshine and Sherrington, is students checking their work for themselves.

We see from the above that independent learning is important to effective reviewing. The best reviewing methods involve students avoiding prompts, such as cue cards or open books, and require students to check and assess their own learning, rather than the teacher assessing students’ work.

Let’s explore further each of Rosenshine’s principles involved in Sherrington’s third strand: (1) daily review and (10) weekly and monthly review.

Principle 1: Daily review

The purpose of daily review is to support the level of fluency a student has with weekly and monthly review (Sherrington, 35). The benefit of daily review is that it ‘allows students to re-activate recently acquired knowledge, reducing cognitive load at the beginning of a lesson that’s designed to build on this knowledge’ (Sherrington, 35).

‘Cognitive load’ is the term given to the amount of cognitive effort required to perform a task.  A good definition of cognitive load is the following, from Frederick Reif:

‘The cognitive load involved in a task is the cognitive effort (or amount of information processing) required by a person to perform this task. If the cognitive load needed for learning becomes excessive, little or no learning can occur’.[3]

As we saw in an earlier blog post, Rosenshine argues that in order to avoid cognitive overload, teachers should present information in small steps and only proceed to the next step once the previous steps have been mastered.

Sherrington offers the following advice to teachers for daily review practices. First anticipate the fact that students ‘don’t necessary recall recent learning readily’ (Sherrington, 35). We need to make prior learning active in students’ working memory; this is important ‘if we’re going to add more layers of complexity’ to their learning, and thereby develop and consolidate their knowledge and understanding. Sherrington writes that ‘the connections we want to engineer won’t happen otherwise’ – i.e., the connections across the schemata involved in the area of learning a student is reviewing (Sherrington, 35).

Second, Sherrington recommends that any questions we use to recall prior learning should make ‘students explore their understanding’ (Sherrington, 35). We should avoid closed questions, such as ‘Does everyone remember what we learned yesterday?’ (Sherrington, 35). As we saw above, reviewing is best done through pedagogical methods that involve extensive independent work from students, so that students explore their understanding in depth, without prompts.  

Third, daily review helps to consolidate understanding and memories. Sherrington draws attention to research from cognitive science on the cognitive processes involved in understanding and reconsolidating memories, such as the recent work of the cognitive scientist and educational neuroscientist Efrat Furst on the neuroscience and cognitive science behind understanding something.[4] This research suggests that ‘there is a natural time delay factor [in recalling prior learning] that teachers should take account of in their teaching’ (Sherrington, 36). In light of this delay, Sherrington writes that it’s likely ‘for students to experience some short-term confusion and lack of fluency as they encounter new material because we don’t re-wire our brains instantaneously’ (Sherrington, 36). A daily review ‘routine’ can fulfil the function of helping teachers to see whether students have ‘secured the knowledge required in their memory in good enough shape to go build it further’ after some time has passed (Sherrington, 36). As students become more fluent with recall, the time delay reduces.

Principle 10: Weekly and monthly review

Weekly and monthly review has two general aims. First, to ‘ensure that previously learned material is not forgotten’. We cannot avoid the fact that human beings forget things, so this aim is best achieved not by trying to eradicate the rate by which we forget things, but by attenuating ‘the natural rate of forgetting’ (Sherrington, 37). Second, to ‘ensure that, through frequent revisiting of a range of material, students are able to form ever more well-connected networks of ideas’ – in other words, ‘more extensive schemata’ (Sherrington, 37).

Weekly and monthly review has at least two benefits. First, it can help ‘students to learn more information’ (Sherrington, 37). Second, it can help students engage in problem solving more proficiently and successfully. Regular and effective reviewing helps to embed and consolidate what we have learned in long-term memory; this makes it ‘easier to be successful with problem-solving as less space in short-term memory is needed’ (Sherrington, 37). Again, this relates to the concept of cognitive load.

Rosenshine reports that ‘more effective teachers’ – i.e., those teachers whose students made the highest gains in achievement tests (Rosenshine, 12) – ‘routinely engage students in a variety of forms of retrieval practice, recalling and applying previously learned material’ (Sherrington, 37). This relates to the fifth of Sherrington’s guidelines above, ‘vary the diet of teaching and learning methods’, as what underpins this guidance is that employing a variety of teaching and learning methods helps ‘students to explore their schemata in different ways, strengthening future recall’ (Sherrington, 40).

A particularly interesting example Sherrington offers of an effective retrieval practice is learning through the construction of narrative structures, which can serve as effective mnemonics. He gives the example of recalling prior learning about the water cycle, by telling ‘the story of a water molecule, starting in the sea, using the correct terminology for changes of state and using the concept of energy’ (Sherrington, 37-8).

Interrogative questioning is another effective form of reviewing. Appealing to research from cognitive science concerning interrogative questioning, Sherrington writes that this ‘has been shown to have a strong effect on future retention as it forces us to form more coherent schemata’ (Sherrington, 38). Examples of interrogative questions are questions such as ‘How does this happen?’ and ‘Why did that happen?’ (Sherrington, 38).

Sherrington mentions a pedagogical hazard in reviewing. We tend to think, he writes, that ‘we’ve learned something if the information is continually presented to us’ (Sherrington, 38). To avoid this hazard, students need to generate versions of the knowledge to be recalled ‘from memory without looking at the source’ (Sherrington, 38). By doing this, students can learn what they know and don’t know (38). Again, this emphasises the importance of students reviewing independently, without prompts such as cue cards or open books.


[1] Originally published in 2010 by the International Academy of Education; republished in 2012 as ‘Principles of Instruction: Research-based Strategies That All Teachers Should Know’, in American Educator. All Rosenshine page references are to the latter.

[2] Woodbridge: John Catt, 2019. All Sherrington references are to this.

[3] F. Reif, Applying Cognitive Science to Education: Thinking and Learning in Scientific and Other Complex Domains (Cambridge, MA: MIT Press, 2010), 361. On applying cognitive load theory in the classroom, see D. Shibli and R. West, ‘Cognitive Load Theory and its Application in the Classroom’ (Impact, February 2018).

[4] See Efrat Furst, ‘Understanding “Understanding”’ (2019).

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