How do children with special educational needs experience retrieval practice?

Rebekah Gear, Primary Education Lecturer, Nottingham Trent University, UK

The impact and application of cognitive scienceThe study of the human mind, such as the processes of though within educational research is increasingly attracting interest from both practitioners and policymakers alike. It draws our attention to the importance of understanding cognitive load by considering the complexities of working and long-term memory, and has subsequently begun informing both policy and practice within the UK. Retrieval practice, coined as a cognitive science learning strategy (EEF, 2021), is advocated as being able to support memory strength, leading to the long-term retention of key knowledge. Despite an emergence in research considering the impact of retrieval practice, few studies have specifically interrogated this within the context of special educational needs and disabilities (SEND) and, more specifically, shared or detailed these children’s experiences in action. 

My study aims to address this, exploring the application of retrieval practice, situated within the context of primary mathematics and mastering multiplication table facts. The intended outcome was to expose the reality and difficulty of applying approaches from cognitive science theory within diverse and inclusive classroom settings, whilst potentially increasing my understanding of how to improve the retention of multiplication table facts. Furthermore, as retrieval practice is mostly considered a ‘low-stakes’ strategy to support both the remembering and the organising of new knowledge (EEF, 2021, p. 21), I wanted to comprehend what children experienced or alleged ‘low’ or ‘high’ stakes to be. Consequently, there were two main avenues of exploration within this study. Firstly, I wanted to observe first-hand whether or not children with SEND perceived a benefit of this type of cognitive science strategy, in terms of helping them to learn and remember their multiplication tables, and secondly, whether in fact they experienced it as the discourse describes, as ‘low-stakes’. 

Retrieval practice theory 

Arguably, at the crux of children’s ability to retain key knowledge is the practice of remembering, referred to as the core of what effective learning is built on (Agarwal and Bain, 2019). Retrieval practice is known as a strategy that seeks to create a stronger retention of knowledge by increasing storage strength (EEF, 2021). Through the process of recalling and rehearsing new knowledge, retrieval practice pulls information out of our long-term memory (recall) into our working memory, strengthening the memory trace (Agarwal and Bain, 2019). Kirschner et al. (2006, p. 77) describe learning as a change in the long-term memory, stating that ‘if nothing has changed, nothing has been learned’. They argue that such a process is a stark contrast to simply cramming information into children’s heads, which can perpetuate working memory overload. 

Despite having previously been defined as the testing effect (Jones, 2019), Agarwal and Bain (2019) and Roediger and Karpickle (2006) stress that retrieval practice is not an assessment tool. Within classroom practices, Agarwal and Bain (2019) and, more recently, the EEF (2021) discussed the different forms that retrieval practice can take, the most common being quizzing. They both consider it to be a ‘low-stakes’ opportunity to support both the remembering and the organising of new knowledge (EEF, 2021, p. 21). Agarwal et al. (2014) previously defined ‘low-stakes’ in this context as meaning that the outcome of the test or quiz does not count towards children’s grades or assessment data. However, there appeared to be a lack of evidence that represented what children perceived either low or high stakes to be.

There is an abundance of research that has indicated both the impact and the success of retrieval practice in action. Adesope et al. (2017) describe how approaches such as tests, including multiple-choice quizzing, have repeatedly provided researchers with evidence of a dramatic increase in the long-term memory’s capacity for retaining and remembering new knowledge. Specific to mathematics, research from Agarwal et al. (2014) which was conducted across several classroom settings, offered evidence that retrieval practice strategies, including quizzes, increased children’s outcomes. Their study further showed the potential of retrieval practice to increase both confidence and engagement, as well as boosting learning for diverse student groups, including those with SEND. 

Nevertheless, as retrieval practice applies to the process of memorisation (Agarwal and Roediger, 2018), in the context of SEND there are a multitude of complexities – not just regarding individual memory capacity – that must be considered. For example, factors, such as children’s self-esteem and identity of their own ability may influence the effectiveness of the learning process (Moreno, 2010). However, previous research directly studying the effects of retrieval practice in the context of SEND claims that brief in-class quizzes boost the learning for these children (Agarwal et al., 2012). More recently, Agarwal et al. (2017) discovered that children with lower working memory capacities benefited most from retrieval practice strategies, compared with their peers with higher working memory. However, both studies reflected on a need for more research to be conducted in partnership with practitioners inside authentic, diverse classroom settings, such as that within which my research study was situated. 

The research

The retrieval practice research intervention was designed to be a follow-up task, leading on from habitual taught multiplication table sessions. Children were issued with a low-stakes quiz to complete the following day. The research gathered evidence from a specific focus group consisting of six children in Year 5 who were on the school’s SEND register, noted as having a moderate learning barrier – a sub-category within cognition and learning, one of the four primary learning barriers identified by the 2015 SEND Code of Practice (DfEDepartment for Education - a ministerial department responsi/DoH, 2015). For context, national statistics show that SENDs are most prevalently recorded in pupils aged 10, reaching a peak of 19 per cent within this age group (DfE, 2021). This study intended to address gaps within the discourse, particularly looking to notice whether children’s own identity and self-esteem created barriers to their experience of retrieval practice. Therefore, a qualitative methodology involving observations and interviews was adopted to support children to share both their views and their experiences. Initial observations were situated to provide a documentary of how children appeared to experience the retrieval practice intervention in action, and a focus group was established to facilitate further exploration of their opinions and perceptions. In line with this study’s context and the needs of the pupil participants, there were extensive ethical considerations in line with both the school and the University’s guidelines, including protecting the children’s identities, using pseudonyms throughout the study. Furthermore, in an attempt to reduce any stress for the children in advance of the research, discussions were had around the purpose of the low-stakes quiz being implemented. It was affirmed that the outcome was not to be used or recorded to support assessment information; it was only to facilitate the children’s personal memorisation of their multiplication tables. 

Two main themes emerged from the evidence gathered, thus forming the basis for discussion and dissemination. These themes challenged the notion of ‘low stakes’ and positioned children’s self-esteem and identity as pivotal. A small selection of the documentary analysis, taken from both the observations and the focus group, is drawn upon next to support the dissemination and the discussion of these two themes.

Low-stakes?

Throughout the literature, the application of retrieval practice is described as forming what is termed as quick, low-stakes quizzing and defined as being depicted as a ‘test’. However, evidence from the focus group discussions provided a stark contrast to this. When children expressed their experiences of the retrieval practice session, they immediately referred to it being a test-like experience and therefore, to an extent, experiencing it personally as ‘high-stakes’:

It’s too much like a test.

I made mistakes in the test because it was too hard.

It didn’t help me remember my times tables. It just made me sad, because I knew I was getting some wrong on the test, and it made me panic!

Children insisted that it did not help them to ‘remember’ their multiplication table facts, even describing how it evoked a state of ‘panic’. This was a significant piece of evidence, particularly as my retrieval practice intervention had been designed deliberately to be, as I perceived, a low-stakes quiz. Yet the children’s commentaries suggested that they had experienced what could be argued as a high-stakes testing experience. This further correlated with evidence from the unstructured observations of retrieval practice in action. During the retrieval practice session, there were indications of stress observed, in a multitude of ways. Arguably, the most obvious example of this was illustrated when one child cried during the intervention, commenting afterwards within the focus group: ‘I can’t do tests, they make my brain forget things!’

Furthermore, it was observed how children’s body language changed during the retrieval practice sessions, with some refusing entirely to complete the task.

The impact of self-esteem and identity

Moreno (2010) argued that retrieval practice theory does not consider other factors, such as children’s perceptions of identity in terms of their own ability, and how these could influence the effectiveness of retrieval practice in action. This was captured when some children shared perceptions of their identity in terms of ability, associating the retrieval practice intervention again with previous testing experiences: 

I couldn’t do it because it was a test and I get the lowest score on tests, because I am stupid and find maths hard.

I am not good at tests; I always get low marks.

This evidence directly from the voices of the children exposed the significance of individual perceptions of identity and how these impacted upon how children with SEND individually both experienced and perceived retrieval practice. This could have been fortified by their own personal association with this experience being a test, and how this experience had for them been ‘high’ and not ‘low’ stakes. These extracts simply scratched the surface of what could truly be the implications of self-esteem and identity for children with SEND within this field of research.

Conclusions

This research illuminated children with SEND’s perceptions and experiences of retrieval practice in action. It exposed a gap within an ever-expanding enquiry of educational interest, placing a spotlight on why classroom strategies, of any nature, need to be carefully implemented and not without listening carefully to children’s experiences and perceptions in action. Perhaps the main message from this study is how vital research is and how important listening to children is to help to facilitate their learning. The use of research methods such as pupil focus groups, as this study proved, is fantastic at enabling this. They can enhance and deepen an understanding of our own perceptions, often informed by our observations of children in action, something that practitioners habitually do in their practice.

In the context of this study, the direct consequence of listening and reflecting upon what these children shared with me within the focus groups enabled me to make adaptations to this strategy and make it work for this particular classroom and its children. A direct implication of this meant that the research later explored other retrieval practice strategy structures, deploying follow-up prompts in the form of simple mathematical games. One example that was particularly effective involved using playing cards, with children co-designing this particular intervention. It was collectively decided that the children could work in partners, flipping over pairs of cards to create multiplication table calculations. The children were able to remove certain card values beforehand to allow them to focus on specific multiplication tables. Not only was this process viewed as ‘low-stakes’ by the children, but it also ignited metacognitive conversations about which tables they wanted to focus on and why.

This study has its limitations, particularly in terms of its very small-scale nature and pupil focus group, being situated in a single year group and classroom. However, I believe that this study could open the doors to future research to interrogate other cognitive science strategies in the context of SEND, and possibly even other student groups. What is certain is that there remains much to be understood in terms of both the application and the integration of cognitive science strategies within our classrooms and practices.