Friday, May 09, 2008

10 facts about learning that are scientifically proven and interesting for teachers

In comments after my post ‘BBCs Paxman Demolishes Brain Gym’ I was challenged to provide 10 facts about learning that are scientifically proven and interesting for teachers. The problem I had was whittling it down to ten!

So here goes:

1. Spaced practice
Perhaps the most significant fact we know about learning, yet it is almost completely ignored by the 'curse of the course and classroom'. We learn through practice, little and often. Ebbinghaus proved it in 1885, and almost everyone in the learning profession has studiously ignored it for well over a century. Demster reported this sad state of affairs in American Psychologist (The Spacing Effect: A Case Study in the Failure to Apply the Results of Psychological Research, 1988). We forget things quickly and that the most effective way to prevent this forgetting is to practice at spaced intervals over time. Knowledge is easy to learn but hard to retain. Forget this and you condemn yourself to, at best to unnecessary effort in learning, at worst failing to learn much at all – the true story behind most learning effort.

2. Cognitive overload
This well know phenomenon is extremely common in teaching and training. A lack of understanding about how memory works leads to a lack of preparation of material in terms of size, order and engagement, leading to weak encoding, a lack of deep processing then poor retention and recall. Almost all courses are too long, present material in the wrong way and lead to unnecessary forgetting. Simplify to prevent cognitive overload.

3. Chunking
Perhaps the easiest and simplest piece of learning theory to put into practice. Chunking means being sensitive to the limitation of working memory. Less is more in learning and distilling, rather than enhancing, elaborating and creating lots of distracting noise, is a virtue in teaching. Unfortunately the ‘song and dance’ act in the classroom is often cacophonous.

4. Order
The order you learn things is critical to how they will be stored and recalled, yet education and training continues to jumble and confuse content. This is critical in language learning, science, maths and indeed, every subject. Learn things in the wrong order and you’ll end up having to unlearn.

5. Episodic and semantic memory
Once you understand that the things we learn are stored differently, i.e. we have different types of memory, then you’ll be more sensitive to the necessary differences in teaching. We still have far too much reliance on text (semantic) for subjects that need a visual (episodic) approach. You see this everywhere, from text heavy PowerPoints to whiteboards, manuals and hand-outs.

6. Psychological attention
Learning does not take place without psychological attention, so setting up classrooms and scenarios that inhibit attention, or distract from learning, is massively counter-productive. I fear that much so called ‘collaborative learning’ falls into this trap. Cramming 30 plus teenagers into a small, airless classroom is no way to encourage attention. There are at least 30 other human distractions, the windows and daydreaming to content with. The bottom line is that most learning is best done on your own or one-to-one.

7. Context
We know that recall is enhanced by learning in the physical context in which one is expected to perform. Yet most teaching is done in alien environments – classrooms ad training centres. We have plenty of proof that work-placed learning needs to be massively increased and non-contextual classroom teaching decreased.

8. Learn by doing
From William James and John Dewy through to Kolb and Schank, we’ve had a torrent of theory showing that we learn lots by doing, yet much teaching and training is locked into a over-theoretical, knowledge and not skills, model. There is a barely a subject around in schools ad training that wouldn’t benefit from a boost in experiential learning.

9. Understand ‘peer’ groups
The work of Judith Harris (The Nurture Assumption) will change the whole way you look at parenting and teaching. Her revolutioary scientific work showed that most books on parenting and teaching overestimate the influence of parents and teachers, and under-estimate the role of genetics and peer pressure. There are some real and practical steps one can take to avoid the obvious traps. These are largely ignored in education and training. Read the book.

10. Murder the myths
This is perhaps the most useful piece of scientific advice for teachers and trainers – dump the snakeoil techniques. These include learning styles, playing music while you learn, Brain Gym, left-right brain theories, NLP, stating the objectives at the start of a course…the list goes on.

Many teaching practices are in direct opposition to the psychology of learning. When it comes to education and training, the professions have doggedly chosen unproven pedagogy over prove psychology. This is why so little progress has been made, and why huge amounts of extra funding leads to such razor thin, marginal improvement. There are literally dozes of proven findings in the science of experimental psychology that are largely ignored. This is what the Bristol study I referred to in my Paxman piece is so worrying.


Mark Frank said...


Thanks for this. It was my comment so I guess I should respond. Let me emphasise this was a genuine enquiry. Your response makes me realise my question was badly phrased. What I was really looking for was a bit of science that would surprise me.

I can explain it better by putting it in context. From time to time I run instructor training courses. Initially I based these courses on what I knew to be best practice among the many people I have come across in the industry. Later I decided to try and put the course on a more scientific footing so I started studying people such as Alan Baddeley and used some of these results in the course.

But I discovered that the science made little practical difference. Almost any one who has been involved in training for even a short time or just reflects on their own experience as a learner recognises that:

* if you don't recall stuff you will quickly forget it
* you can't absorb too much at one time
* you learn by doing
* you need to avoid distractions
* you need to structure content logically
* you should be wary of too much text

The fact there is underlying science to support this best practice is somewhat comforting but makes little difference to what is already recognised as the right thing to do. And of course I want instructors to learn by doing and not immerse them in theory.

I think the interesting question is not "what is best practice?" - but "why best practice is not done more often?".

Perhaps the value of the science is to help resist the pressures to do things badly?

Donald Clark said...

Thanks mark - I'm not so sure about the observation#;

"The fact there is underlying science to support this best practice is somewhat comforting but makes little difference to what is already recognised as the right thing to do."

In my experience, God is in the detail here. A detailed understanding of these points leads to changes in practice that have a real pay back. On spaced practice you really do need to understand more than the general principle (although even that helps enormously). I have seen no evidence that practising teachers and trainers, even when they know of the principle, really apply it in practice. This is because they're locked into old classroom and course structures.

Have a look at the homework (if you get much) that is set by teachers in schools. have a look at the structure of most corporate courses. Mistakes are made because people ignore the basic theory. Most of the cotet is too log, wrongly ordered, in the wrong medium and so on.

I have yet to come across a teacher or trainer who has read Judith Harris or is aware of the details of her work.

On the other hand almost every encounter I have with teaching and training shows me that they do believe in old, outdated and discredited theory.

Mark Frank said...


I am confining my comments to commercial adult training. I don't have enough experience of primary or secondary education.

In my experience, God is in the detail here. A detailed understanding of these points leads to changes in practice that have a real pay back. On spaced practice you really do need to understand more than the general principle (although even that helps enormously).

I challenge how useful the details are in practice. Take spaced practice as an example. I take it that the principle you are referring to is to have practice spaced out over increasing intervals after learning. I remember having this pointed out to me on an instructor training course about 30 years ago. It is a sound idea, I was delighted to find that the science backs it up, and I try to implement where I can. But I haven't found any more useful detail beyond that. For example, I am not aware of a table that allows you look up the appropriate spacing intervals for a real context. If there were such a table would you trust it in preference to an experienced trainer's judgement?

I have seen no evidence that practising teachers and trainers, even when they know of the principle, really apply it in practice. This is because they're locked into old classroom and course structures.

I agree the first sentence is often true, but I think you are too hasty with the second sentence. In my experience poor practice is much more to do with the constraints placed on the trainer professional who depressingly knows it is not good but hasn't the power or skill to change it. For example, I have just come off a telephone call with a lady responsible for an IT roll-out discussing the end-user training. She was concerned that not everything that users needed to know was covered in the 60 minutes Powerpoint presentation. She acknowledged that the users would not absorb this extra information, but that wasn't the issue. She needed to show that if something went wrong the users "had the information" so it was their fault not hers. In my experience this is much more the kind of thing that leads to poor practice. Maybe what is needed is better negotiation skills for trainers.

Rob said...

Great stuff. I can't disagree with anything here. I was reminded, though, that several things you list here - about recall and learning- I first encountered in a book called "Memory" by IML Hunter, on my PGCE course thirty (blimey!) years ago. I read it because I had to - we had to learn about how people learn. Such a text would never be on a PGCE reading list now.

Doug Holton said...

Here's a related list, the word file at the bottom of the page has more details:

Mark Frank said...


A really useful list thanks. It is very valuable as a kind of checklist. I still think that almost everything in there would be something that an experienced instructor would already recognise as good practice - but it is useful to be reminded and see the academic backing.

What is most valuable to me is that a list like this clearly separates sound advice from snake oil.

Donald Clark said...

On spaced practice there's lots more detail.

Before starting, the fost poit is to avoid cogitive overload in the first place ad get the meaterial to be learnt into a chunked and memorable form.

1. Robert Bjork and Thomas Landauer discovered that the best time to reiforce is when we're about to forget (studies with 700 undergraduates).

2. The forgetting curve isn't any old curve. It drops dramatically, then levels off. This means that the spacing needs to be less and less frequent as one goes forward in time.

3. The first reinforcement is best left for ten minutes to allow for some consolidation. It is then useful to reinforce within hours on teh same day, also usueful to do so before one goes to sleep, then the nenxt day and at less frequent intervals spacing out to weeks. A 10 minute, two hour, 1 day, 2 day, 10 day, 30 day, 60 day pattern should work. It sound like a lot but it does work.

4. In fact, with spaced practice there are pieces of software that do the job for you e.g. SuperMemo. You can also set up reminders in whatever Calendar, Memo system you use on your PC.

Donald Clark said...

There's also these references (from Doug's link:

Bahrick, H.P., Bahrick, L.E., Bahrick, A.S., & Bahrick, P.E. (1993). Maintenance of foreign language vocabulary and the spacing effect. Psychological Science, 4, 316-321.

Cepeda, N. J., Pashler, H., Vul, E., Wixted, J. T.& Rohrer, D. (2006). Distributed practice in verbal recall tasks: A review and quantitative synthesis. Psychological Bulletin, 132, 354-380.

Cull, W. L. (2000). Untangling the benefits of multiple study opportunities and repeated testing for cured recall. Applied Cognitive Psychology, 14, 215-235.

Donald Clark said...

Ad another thing! There are two studies that suggest that 'writing' in your own words durring learing will significantly increase recall. I'm amazed when I give talks at coferences to learig professionals and most of them have othig to write with, or don't take notes. It is, in itself, an immeduate form of speced practice ad results in deeper cogitive processing. These otes ca then be polished and used for spaced practice. I do this with most books I read - underline, take otes, read notes again.

Donald Clark said...

I do think that teaching and training, as currently practised, would be radically altered if we took this 'spaced practice' theory seriously. There would be more atttention paid to reinforcement by the learner, than straight teaching. School timetables would take note of the need, studets could be give practice schedules etc. Software could be used to remind and reinforce.

Training courses are worse than schools, because there is one of the scheduled classes over time to reiforce. However, the compter is the natural prompt.

Language learing, in particular needs to be completely rethought in schools around this principle, and others.

How much better would this be than Brain Gym and other fashionable nonsense?

Mark Frank said...


Forgive me pushing this a bit, but I am surprised by your confidence in the correct timings for rehearsal. Here are a couple of quotes from the Cepeda et al reference (which is a 2006 literature review of the subject).

In summary, quantitative syntheses of the temporal distribution of practice literature have suggested that a benefit from longer ISIs is a fairly robust effect. Beyond that, however, few firm conclusions
seem warranted.

(ISI=interstudy interval)

Although the literature on distributed practice is indeed very large, the present review discloses (in ways that previous reviews have not) how sorely lacking it is in the very sorts of information that are most needed if serious practical benefits are to be derived from this century-long research tradition.

Where did you get your timings from? How did the studies demonstrate that these timings were applicable to all subject matters? The Capeda paper appears to say that results vary greatly with subject matter. For example, talking of an earlier literature review:

In contrast, only one third of intellectual skill (e.g., math computation) studies showed a benefit from distributed practice, and half showed no effect from distributed practice.

Maybe these are not the best references?

Unknown said...

Don -- Love your site! And agree with 99% of its material as we've been saying the same things for years in IBM Learning...

Two "nits" however:

1. Nothing is ever "scientifically proven." Good science can support, strongly suggest, but never "proves."

2. "Episodic memory" refers to retention using story schema. (Episodic means episode). While visual salience may be included in that, the operant element in episodic memory is "story-based content" as differentiated from lists or item-based.

Keep on carrying the torch for evidence-based learning design!

Thank you.

Peter Orton

Anonymous said...

I noted a school in Tyneside (where it all happens as you well know!) is trialling the spaced practice method. Paul Kelly, the local headmaster and his staff are pushing this outide of national remits (constraints?).

Article link:

Professor Alan Smithers of Buckingham University sounded a 'note of caution'. I guess its his job to.

Martin M-B said...

Stephen Downes comments in Stephen Downes style...

Donald Clark said...

Stephen's makes two points:
1. Be careful about the phrase 'scientifically proven'. This is fair. The problem, i my view is that much for what passes as theory i learnig has ever been tested with any scientific rigour, discredited by science and even sold as 'science' without any actual citations (NLP, learning styles etc). I'm careful in that I like to adhere to science as falsifiable hypotheses and a methodology that includes experimetal evidence. The absece of these is my watchword.

2.Claims that '3 & 8 are inconsistent' 'Chunking' and 'learnig by doing'. I don't get this and neither do the subsequent commentators on Stephen's blog. The presentation and structure of content and tasks (chunking to avoid problems with cogitive overload and facilitate storage ad recall) is entirely consistent with the 'practice' and 'application'.

Anonymous said...

Reading these posts I am re-considering this career as ID. This is so beautiful, my calling to read and to research. Should I drop the job and do a PhD? Sometimes small pointers lead you to right direction. No words to say how wonderfully you have put the things. The observation about practice is what I always wondered. You can't ever realis how much you have helped me by sharing this knowledge.
1. I was having problems retaining some new things that I have picked up. Now I know that there can be a learning curve and we need to reinforce the learning at certain gaps. Without this information I was blindly trying to grasp what was wrong with my memory.

2. My daughter has been writing certain alphabets facing the wrong direction, reversed. Sometimes she writes her name like a mirror image. This thing I had not faced with my son. Now it is obvious, I was making him practice and this had perfected his understanding. I have not been able to give enough time and attention to her and then I was wondering if she was having some learining prolems. This post cleared the doubts. It is the practice, which has been lacking.

How can I thank you, This is a blessing. You write beautifully and this is one of the most well-crafted posts I have ever read. This seems to be just the tip of iceberg, I wish I had discovered these amazing works earlier. Accept my gratitude and I will be reading all these posts.

Anonymous said...

I really enjoyed reading your blog. very well written. Could you please explain point number 10. murder the myths, if it is not too much trouble?

CandaceID said...

I, too, would like some clarification on #10. Using just learning styles as an example... Contrary to your preference (& my previous 20 years of practice), for past 5 years I have studiously avoided taking notes during workshops & meetings. I discovered that it has HELPED me to better attend to, understand, AND remember more. I believe I must have some type of hearing disability (very visual learner). I can hear something a dozen times & not get it, but if you breeze the page past my nose a few seconds, I can ace your test. I positively KNOW that my children have different "aptitudes" and learning styles and research supports this. I also KNOW that when I make the effort to address as many styles as possible even in small ways, that my trainees & grad students all perform better. This of course is due in part to additional repetition/receiving info in different modes, but still supports consideration for such "snake oil." Let's not forget that snake oil may actually help some folks -- like those with dry skin, even if it doesn't cure every thing for every one. ;)

Donald Clark said...

My views on Learning Stles are not based on my (25) years of experience but a serious scientific examination of these theories (there are many of them) by people such as Professor Frank Coffield, Susan Greenfield and many others. Serious scientists and thorists don't take 'Learning Styles' theories seriously,as they are simplistic and pigeon-hole learners. See my other posts on the subject.

Your experience with note taking is interesting, but perhaps an example of a distribution around a mean. there are solid studies showing that note taking (in your own words) has a dramatic effect on retention. Note that a scientific study with this claim, doesn't imply that it has this effect on everyone, only that it is statistically true for a tested sample with a measured distribution. The fact that you're different doesn't invalidate the claim. Interestingly, those who are good at elaboration, first explored by Craik and Lochart, where you encode by elaborating (the meaning, visual or other cues) may achieve good retention without taking notes, but few have the ability to do this from working memory for a sustained period of time, hence note taking. The science says this also.

Millard said...


Interested layperson here. I'm writing an article on "proof" and was looking for examples of things people say are "scientifically proven." I came across several links to your interesting article. It turned out to be a portal to ISD, which I'm now reading up on.

A couple of questions to help me understand. What in your mind constitutes "scientifically proven?" And specifically about this list of 10 items, how were they proven--by virtue of specific studies, longtime practice, consensus among successful trainers, none of the above?



Donald Clark said...

By 'proof' I mean scientific proof through either observation or experiment - but with rigour in the methodology.

Mark Frank said...


This is a subject which interests me greatly and which I have contnued to follow in the year since Donald's initial post.

Donald talks of scientific proof and rigour in the methodology, but in the social sciences and psychology there tends to be a trade off between rigour and usefulness.

For example, look at the discussion about spacing and memory between Donald and myself above. As Donald says, there is ample evidence that spaced repetition reinforces memory - but anyone who has to remember a poem or the lines from a play would recognise this - it hardly needs an experiment. What would be useful would be evidence on how much spacing to use in real life situations. But the studies on that are inconclusive - see the Cepeda reference above.

I have had to read a number of social science/psychology papers recently and so many of them suffer from physics envy. They give the outward appearance of being both rigorous and deep by including control groups, statistical tests and concluding with the dreaded p value. But they deserve a sceptical assessment almost as much as the straightforward quackery of the brain gym or whatever. It is very rare for them to combine rigour and depth and often they lack both!

Unknown said...

I've used the spaced-repetition software Anki for about a year now for language learning and IT certification studies. It's amazing how much it helps me retain information. Before using Anki, I would learn a word or phrase, then forget if I didn't use it for some time. I was frustrated because I knew that I knew the information, but simply couldn't recall it. Anki is totally free. It has changed how I study, dramatically, for the better.

Sorry if I sound like a salesman...I'm just a huge fan from the results I've had with it.

Also, here's a great article from SuperMemo. Effective learning: Twenty rules of formulating knowledge:

Fiona Weir said...

Interesting 'facts' but I really would like sources/references for all of them so I can evaluate the evidence for myself, please!

Anonymous said...

Thanks for this, very interesting reading, refrences would make it better but it looks like these may be in the comments.
Since you suggest "Murder the myths" perhaps a list of the ones most in need of murdering would be useful too. I am particularly interested in your comment that giving objectives is not useful.

Unknown said...

Isn't episodic memory more around narrative? Build a story/event and more likely to remember? Not necessarily visual?

Roy Wright said...

Could you elaborate on what you said about stating the objectives? I'm not familiar with what you're referring to, but it sounds intriguing.

Donald Clark said...

7 reasons: Why we need to kill boring ‘learning objectives’!