What is the myth of Dale’s Cone of Experience?

Although not as widespread as some of the other learning myths, if you put the search term ‘Dale’s Cone of Experience’ into google you receive a myriad of pages supporting it. These even include pages from university courses! Typically they contain an image like the one below:

The Cone appears to show that the way in which something is presented impacts on how well you remember it. It seems obvious from looking at the cone that reading is not a good way to remember anything. And that the best way to teach something is to by ‘Doing the Real Thing’ as then apparently you will remember 90% of what you say and do! Unfortunately, just a small amount of actually thinking this through will help you to realise that this just isn’t true. Can you really remember exactly which pair of socks you put on 2 weeks ago? Or what you were doing at 2 pm last Tuesday? Or even the last thing you said yesterday? If you were to remember 90% of what you say and do then these questions should be easy for you to answer, but my guess would be that actually they weren’t. Whereas you might actually do better if you were asked for something interesting you read in the news last week or the plot of an exciting story (even if you read it many months or even years ago).

This focus on remembering more of what we say and do has led to the idea of ‘discovery-based learning’. In Canada (Crehan, 2016) they moved the maths syllabus towards a more discovery-based learning approach. The idea was to encourage students to develop their problem-solving skills by getting them to discover how to solve mathematical problems for themselves. If Dale’s cone was correct this should produce the best retention of mathematics possible. Unfortunately, it’s not true and the results from Canada highlight this – a report by Anna Stokke (Crehan, 2016) showed that student’s performance in provinces that had moved to a discovery-based approach dropped. Their performance on fraction questions dropped to only slightly better than chance (i.e. if they just guessed the answers)! In a meta-analysis of 285 studies, Hattie (2008) found only a small effect (0.15) for discovery-based learning which doesn’t suggest the big effect of retention in the cone above!

Beyond the outcomes being untrue, after looking through the images produced by a Google search, you begin to see some inconsistencies in the figures and the order of the learning experiences. If you carry out a little research into the history of the pyramid,you realise that someone probably made up the numbers (Lalley & Miller, 2007) – which explains why they are such ’round values’ (wouldn’t it strike you as surprising if psychology research really gave exact values for learning of ‘10%, 20%, 30%’ etc?). In fact, the idea of Dale’s cone has been corrupted from his original position nearly beyond recognition.


So what did Dale actually mean?

Dale (1946) wasn’t trying to show how much we learn and remember. Instead, his pyramid was designed to show how learning progressed and occurred at each level from concrete to the abstract. Dale saw the cone more as a continuum rather than as having discrete categories. He recognised the importance of having a variety of learning experiences (Lalley & Miller, 2007). Don’t forget that in the 1940s education was quite different from the present day – Dale was trying to make the point that some learning experiences were better for learning certain things. For instance, if you sit and watch a chemical reaction you will not get the most of the learning experience. It would be a far more meaningful experience if you knew which chemicals were present and understood a bit about how those chemical are likely to react and why. Dale recognised that learning has many routes and that teachers (who have experience of teaching certain topics) were the best placed to pick these routes.


What evidence is there that we learn better from different types of media?

The other reason to be dubious of the figures usually associated with Dale’s Cone of Experience comes from the difficulty of actually generating the figures that are being claimed. Having run lots of studies in psychology we can tell you that the numbers never come out that neat! Even if, by fluke, they had – how would the study have to be run?

In the example cone above there are 6 different conditions. For each condition we would need to create teaching materials which covered the same information, took students the same time to absorb and there wasn’t an obvious bias in the way they are best presented (a task that would be challenging to say the least, if not impossible). The best way would be to have six different topics and different groups of students where each topic came in a different format. We would then need a delay before testing (as we’re studying retention). We would need to control aspects of our population – age, ability etc. And we would probably need a very large sample (>1000) to get any meaningful data. In short, this is a big project and, as far as we can tell this study has never been run.

Lalley & Miller (2007) did a review on the evidence for learning from each level of the cone. They found that each of the methods mentioned in the cone is effective for retention. But that there is no evidence for the superiority of one over another. It is likely to depend on what is being taught.

Dale’s original conclusion seems to stand. Using a variety of appropriate teaching methods is the most effective way to teach, and it is still the case that the best way to decide which teaching method is to ask a teacher! There are so many factors which go into ‘the best way’ to teach something that it is hard to come up with a checklist which would work in every situation. This is why Get My Grades has used the experience of a multitude of teachers to create our learn pages and aims to use videos and images where appropriate alongside clear, written explanations. There’s no magic learning strategy – the version of Dale’s Cone that is commonly displayed, even on teacher training courses, is misleading at best.



Crehan, L. (2016) Cleverlands: The Secret Behind the Success of the Worlds Education Superpowers. Unbound.

Dale, E. (1946). The cone of experience. E. Dale, Audio-visual methods in teaching, 37-52.

Hattie, J. (2008). Visible learning: A synthesis of over 800 meta-analyses relating to achievement. routledge.

Lalley, J. P., & Miller, R. H. (2007). The Learning Pyramid: Does It Point Teachers in the Right Direction?. Education, 128(1), 64-79.