Thread by @DrRoseLea

Our @EducEndowFoundn Cognitive Science in the Classroom Review is out! There is a full report and a shorter, more digestible summary report. Last week’s #CogSci preview talk from @TWPerry1 here: tinyurl.com/4bxu8tw5. I’ll attempt to summarise below... 🧵 (1/25)

https://twitter.com/EducEndowFoundn/status/1415920394536620033

Cognitive science is increasingly used to inform interventions, practice, and policy in education, and underpins many classroom practices (e.g. spaced practice, dual coding). It forms a key part of Ofsted’s Inspection Framework & often teacher training (2/25)

We & @EducEndowFoundn wanted to know whether #cogsci techniques work in real classrooms, across the curriculum, for diff pupil groups. We completed a systematic review of classroom trials & practice review (inc a teacher questionnaire & in-depth interviews) (3/25)

Using a preregistered protocol, we searched 10 databases for all classroom trials that had tested a cognitive-science informed strategy. We systematically sifted through 40,000+ (!) papers, down to ~500: with 43 ‘high priority’, 252 ‘medium priority’, 204 ‘low priority’ (4/25)

High/med papers analysed. These met several criteria: examined impact of a cogsci-informed strategy on learning outcome(s) for children & YP aged 3-18, had comparison/control, & used real school setting (undergrad/lab studies excluded) (5/25)

Categorising and grouping this diverse group of studies was challenging. We divided our studies into 14 strategies within 8 areas (see pic). Each area is summarised briefly(!) below... (6/25)

1. Spaced practice (spreading out learning/teaching of material over time) seems to improve learning for most pupils in most subjects. Little evidence on within-lesson spacing, but promise. Can be challenging to implement due to how curricula usually organised (blocked) (7/25)

2. Interleaving (learning material interspersed with different but related content). Moderately +ve effects, but strong focus on maths (1 in PE), for children 8-14. Some teachers concerned interleaving may confuse pupils with lower attainment, who like compartmentalising (8/25)

3. Retrieval practice (e.g. quizzing) studies spanned ages from EY to age 17, with good subject coverage, inc languages, history, geography, maths, science. Moderate +ve effects, but some caveats & less clear for more complex/subtle content (see review) (9/25)

4a. Strategies that manage cog load aim to avoid working memory overload. Worked examples can have +ve impacts on learning, but studies focussed on maths & science, mostly with secondary pupils. Using incorrect or partial worked egs needs to bear in mind prior knowledge (10/25)

4b. Cog load can also be managed thru scaffolding- guidance, prompts, etc. Well-targeted scaffolds can aid problem-solving in KS2-KS4 in multiple subjects, but need to account for prior knowledge. Optimising cog load via group work is possible but challenging to get right (11/25)

5. Working with schemas (mental structures that organise knowledge)- Evidence on concept-mapping (e.g. organisers, mindmaps) is mixed – depends on many factors (e.g. who generates, how pupils engage, how implemented). Evidence also mixed re: comparison and conflict (12/25)

6a. Cog theory of multimedia learning/dual coding (presenting info in multiple formats) – evidence mixed. Visual representations can be effective, but they can also increase cog load if used incorrectly. Informational images generally more effective than decorative (13/25)

6b. Diagrams (providing info schematically) – sometimes positive effects on learning for secondary science/maths. Small-moderate +ve effects of spatial/visualisation/simulation approaches for primary pupils, but heavy focus on maths (e.g. mental abacus) (14/25)

7. Embodied learning (using movement & body to support learning)– focussed on primary aged children, but included a range of subjects (maths, science, reading, vocab). Considerable promise but little evidence. +ve results for use of gesture, play & tracing for learning (15/25)

8. Mixed strategy programme trials (with >1 cog sci strategy) were amongst the strongest egs of applied research (ecologically valid & at scale). Often involved teacher CPD & school-level implementation. Evidence mixed - successful implementation is difficult (16/25)

Our overarching finding is that there is value in teachers having working knowledge of cognitive science principles. Applying #CogSci principles can have real impact on classroom learning. However… (17/25)

There were notable gaps/complexity/uncertainties. Most #CogSci classroom studies were in maths and science (though English, EFL, and humanities made appearances). Relatively few with primary school children and EY (18/25)

Another caveat - while all reviewed studies took place in classrooms, often the lessons were designed and/or delivered by researchers, or teachers given scripted set pieces to deliver. Often trials used researcher-designed outcome assessments (19/25)

The review suggests #CogSci PRINCIPLES are useful, but determining how these should manifest in specific teaching and learning strategies has some way to go. There is unlikely any ‘silver bullet’ strategies that work for all students, subjects and settings (20/25)

Effectiveness of #CogSci strategies likely to depend on many factors… child age, prior knowledge, nature of topic, nature of task, social/emotional factors, teacher-specific factors.. (21/25)

Ultimately, cognitive science matters. Many approaches are promising. But, we need to resist oversimplifying ('evidence says X works everywhere') & instead grapple with ‘messiness’ of real classrooms, & explore relevance to all pupils & subjects. We hope the review helps (22/25)

This project was a team effort by colleagues at Birmingham @unibirmingham @TWPerry (PI), myself @DrRoseLea, Clara Joergenson, CUREE @PhilippaCcuree, Kimron Shapiro @deborahyoudell & BrainCanDo @BrainCanDo @AmyFancourt1 (23/25)

We also had a fantastic diverse Advisory Group, inc headteachers, cog neuroscientists and experts in education research, policy & practice, who provided expertise & guidance throughout the review (see full report for more info) (24/25)

End of 🧵! Please read the full report and summary report here: eef.li/cog-sci