“The mind is at last yielding its secrets to persistent scientific investigation.

We have learned more in the last 25 years about how the mind works

than we did in the preceding 2500”.

Daniel Willingham, 2009.

The more we learn about the brain, the more we learn how much knowledge and memory matters.

Last post, thanks to Ben Goldacre, Tom Bennett and Andrew Old, I explored the difficulty of distinguishing between scientific research and neuro-myths. Things moved since. The Monday after, Tom Bennett launched with astonishing energy into organising the first ever teacher-led wiki-conference on education and research for September. Everything from format, venue, speakers, helpers, sponsors, web design and even its name is being crowd-sourced through social networks. Answering Ben Goldacre’s call to arms, championed by both teaching and research communities, the barricades of evidence-based practice are well and truly manned. Within 24 hours Tom had amassed a mailing list of over 100 people; 24 hours after ResearchED went live it had over 300 followers. Sam Freedman, David Weston and Ben Goldacre were quickly confirmed as speakers. The appetite for this is out there.

This post, I want to set out how scientific enquiry and research evidence is discovering how the mind learns, and might guide us towards ‘working out what works best’- the ResearchED tagline.

All sorts of things strike us as important in learning, given basic physical, physiological and emotional security: extrinsic and intrinsic motivation, expectations, mindset, prior knowledge, intrinsic interest in the topic, perceived relevance, challenge, curiosity, attention, focus, effort, comparing examples, practice, feedback, memory, revision, summarising, choice, self-discipline, responsibility, ownership, parental support, role models, the emotional connection with the teacher and the organisation of the material, hard work … the list could go on and on. Understanding what matters most in learning is crucial if we are to focus our teaching.

Is Cognitive Science good or bad science?

Cognitive science is an interdisciplinary field of academic researchers from psychology, neuroscience, linguistics, philosophy, computer science, and anthropology who seek to understand the mind and apply the findings to education. So, is cognitive science just as much pseudo-science as brain gym, or does it practise what it preaches? In this blog post I want to strip cognitive science down to its essence, and apply two litmus tests:

One. To what extent is the scientific research robust, peer-reviewed and rewarding when re-read?

Two. To what extent does the scientific evidence have practical classroom applications that reward re-using?

 Litmus Test One: Just how robust is the research?

Decades of scientific research (from 1968) have explored what’s vital for learning. In a nutshell, here are three core principles, and their intuitive and empirical grounding.

1. Working memory load should be minimised.

2. Long-term memory retention should be maximised.

3. Knowledge schema should be accumulated and automated.

 1. Working memory load

In a nutshell, working memory is used for thinking, and because it is like a small bottleneck, it is easily overloaded. If working memory is overloaded, it makes new things much harder to learn and remember.

Intuitively, we know this to be true from learning to drive and learning to teach. When starting to drive, there’s too much to keep in mind: pedals, gears, indicators, mirrors, steering and instructions overwhelms us. Not dissimilarly, when starting to teach in your first school, your working memory feels bottlenecked: new systems, policies, rules, 200+ names, locations and sanctions all overload your brain.  Similarly, new Year 7s feel overwhelmed by so many new classmates, teachers, subjects, concepts, older students, rules, locations and playground interactions, and many tears are shed.

Empirically, the past half-century of international research has provided overwhelming and unambiguous evidence on this issue, with Atkinson & Shiffrin (1968) as the first exploration into working- and long-term memory. Meta-syntheses from Sweller (1998), Kirschner et al (2006) and Willingham (2009) are conclusive.

 2. Long-term memory retention

The insight here is that long-term memory is a vast storehouse that helps us overcome bottleneck limitations in our working memory. So learning is actually remembering in disguise: ‘if nothing has been changed in the long-term memory, nothing has been learned.’ (Kirschner et al 2006, p77)

Intuitively, we know this is true from learning to read. Before we could read, letters on the page meant very little. We didn’t have the stores in our long-term memory to decode the symbols, and struggled with what is now automatic. Try memorising the ten symbols ‘%^&$£@&*!@’ as compared to ‘the boys ran’ for an example of the powerful chunking of your long-term memory. Or try 7×8 as opposed to 18×7: one is stored and retained automatically, one requires working memory capacity to add up 10×7 + 8×7 to 126. Similarly, all of us have in the past crammed for exams that if we took now, we’d probably fail. That’s because our long-term memories require usage for retention. The more work is done in retrieving the memory, the stronger it becomes.

Empirically, the critical importance of long-term memory has been established beyond all reasonable doubt:

‘Our understanding of the role of long-term memory in human cognition has altered dramatically over the last few decades. It is no longer seen as a passive repository of discrete, isolated fragments of information that permit us to repeat what we have learned. Nor is it seen only as a component of human cognitive architecture that has merely peripheral influence on complex cognitive processes such as thinking and problem solving. Rather, long-term memory is now viewed as the central, dominant structure of human cognition. Everything we see, hear, and think about is critically dependent on and influenced by our long-term memory’. (Kirschner et al 2006, p76)

3. Knowledge schema

The more knowledge you have, and the more automatically you can access it, the easier you find it to remember new knowledge, and the faster your skills develop.

Intuitively, many teachers know this from teaching mixed-ability classes: those pupils who know more to start with, learn faster.  You also know this from reading an article in The Economist on education compared to finance: your existing knowledge helps you get to grips with what it’s all about. It’d be easier for us to debate from memory the education system in England than the system of head-hunting in Borneo, because we know so much more about one than the other.

Empirically, in studies since 1988 a vast research base has been built up that shows the powerful, beneficial effects of knowledge on memory (Recht et al 1988; Alexander et al 1994; Cummings et al 1999; Van Overschelde 2001, Willingham 2007, 2009 etc).

Given that learning is inhibited by working memory overload, accelerated by long-term memory retrieval, and automated by knowledge accumulation, the imperative for teachers within and across lessons is clear: minimise overload, maximise retrieval and automate and accumulate knowledge. The research is robustly peer-reviewed, and passes my litmus test of rewarding re-reading: the more I look into the cognitive psychology of how the mind learns, the more I get out of it in terms of how to teach. It’s like reading the cheat codes to intelligence. 

Litmus Test Two: Just how practical is the evidence?

My litmus test here is simple: the more you apply it in the classroom, the more useful it should become. Apart from the five core practices of instruction, including examples, questions, practice, feedback and misconceptions, which I have blogged about here, there are three specific teaching tools suggested by cognitive scientists applicable to learning work across subjects and age groups (Sweller et al 1998), (Kirschner et al 2006), (Dunlosky et al).

1. Worked Examples

2. Completion Problems

3. Process Worksheets

Some of the best teachers I know use these without the labels. If you don’t use them, try them out, and see for yourself if they pass the litmus test. Certainly, the more I use them in lessons, the more my pupils get out of them.

1. Worked Examples

The worked example effect, replicated a number of times (1985, 1987, 1992, 1993, 1994 1996 & 1999), shows that learners required to solve problems perform worse on subsequent test problems than learners who study the equivalent worked examples. Studying and comparing lots of worked examples reduces cognitive overload. Working memory is freed entirely for the study of the problem and solution steps. In English, and other subjects with a heavy writing load, this means getting students to compare worked examples of model paragraphs, either to criticise and improve, or to annotate and aspire to. If they haven’t seen an example of what they’re aiming for, how can they work towards achieving it? The best teachers write lots and lots and lots of example paragraphs, introductions, conclusions and essays. In Maths, worked examples are step-by-step model solutions to the problem.

2. Completion Problems

Completion problems are worked examples with partial solutions, where students complete the rest of the solution. Writing frames in English help by preventing cognitive overload and forcing students to make lots of strong analytical points in concise paragraphs. Acronym mnemonics such as PEEL in English and SOHCAHTOA in Maths help students’ retention of the underpinning process in the long-term memory.

3. Process Worksheets

Process worksheets guide students through a sequential series of steps required to solve a complex problem like completing the square in maths or comparing poems in English. They minimise overload whilst maximising retrieval.

So, what can science tell us about how pupils learn? And how can we teach so that pupils learn best? Overall, the scientific evidence is conclusive that pupils learn best by effort and hard work: thinking, concentrating, practising, memorising and recalling subject content. Teachers can accelerate pupils’ learning by minimising cognitive overload, then specifying, sequencing, testing and revisiting subject knowledge until it becomes automated in their long-term memory.

Whether this is pseudo-science or the real deal, Ben Goldacre’s Randomised Controlled Trials and Tom Bennett’s ResearchED conference may yet find out. Let the trials begin!

Research Base:

Dunlosky, J., Rawson, K.A, Marsh, E.J, Nathan, M.J., & Willingham, D.T.: Improving Students’ Learning with Effective Techniques: Promising Directions From Cognitive and Educational Psychology Psychological Science in the Public Interest 14(1) 4–58

Sweller J., van Merrienboer J. and Paas F. (1998). Cognitive Architecture and Instructional Design. Educational Psychology Review, 10(3), pp.251-296

Willingham, D.T. (2007) Cognition: The Thinking Animal. Upper Saddle River, NJ: Prentice Hall. The graduate textbook of cognitive psychology

Alexander, P.A., Kulikowich,, J.M, & Schulze, S.K.: (1994) How subject matter knowledge affects recall and interest. American Educational Research Journal, 31, 313-337.

Rohrer, D. & Pashler, H. (2007) Increasing retention without increasing study time. Current Directions in Psychological Science, 16, 183-186. Distributed practice leads to more enduring memory

Cepeda, N.J, Pashler, H & Vul, E: (2006) Distributed practice in verbal recall tasks: a review and quantatitve synthesis. Pscyhological Bulletin, 132, 354-380. Comprehensive review of the effect of distributed practice on memory

Cumming, J & Elkins:j (1999) Lack of automaticity in the basic addition facts as a characteristic of arithmetic learning problems and instructional needs. Mathemetical Cognition, 5, 149-180. One for the Maths teachers…

Bransford, J.D. Brown, A.L, & Cocking, R.R: (Eds) 2002: How people learn: Brain, mind, experience and school. Washington DC, National Academy Press. Lessons from the science of human learning recommended by Willingham as accessible.

Feldon: Cognitive load and classroom teaching: the double-edged sword of automacity. Educational Psychologist, 42, 123-137.

Recht, D.R: & Leslie, L: (1988): Effect of prior knowledge on good an poor readers’ memory of texthttp://www.mendeley.com/catalog/effect-prior-knowledge-good-poor-readers-memory-text-1/. Journal of Educational Psychology, 80, 16-20. Prior knowledge has a powerful impact on lasting memory.

Van Overschelde, J.P & Healy, AF (2001) Learning of non-domain facts in high- and low-knowledge domains. Journal of Experimental Psychology: Learning, Memory & Cognition.

Kirschner A., Sweller J. and Clark E., 2006. Why Minimal Guidance During Instruction Does Not Work: An Analysis of the Failure of Constructivist, Discovery, Problem-Based, Experiential, and Inquiry-Based Teaching, Educational Psychologist, 41(2), pp.75–86