I like reading Chris Pappas the same way I like eating sour candy – if it doesn’t make me wince, it hasn’t done its job. Fortunately (or not) Chris’s latest article, Cognitive Load Theory and Instructional Design, does the trick. It reminds me of a Dilbert cartoon where the pointy-haired manager says, “In order to be profitable, we need to cut costs and increase sales.” The engineers agree that this is a good idea and ask how exactly are they supposed to do that? The manager replies, “Don’t ask me. I’m an ‘idea’ person.”

True to form, like the manager, this article provides information that is completely accurate and utterly unhelpful.

Chris avers that Cognitive Load theory adheres to principles that should be kept in mind when designing an eLearning (and presumably, traditional) course:

1.       You can reduce the amount of load that is being placed upon the learners’ working memory by integrating the various sources of information, rather than giving them the various sources individually. 

2.       In tasks or lessons that require problem solving skills, avoid using activities that require a “means-ends” approach, as this will place a load upon the working memory. Instead, use goal-free problems or examples to illustrate the point.

3.       Reduce the amount of redundancy in eLearning course design in order to reduce the amount of unnecessary repetition-induced load that is put upon the working memory.

4.       Use visual and auditory instruction techniques to increase the learners’ short term memory capacity, particularly in situations where both types of instruction are required. 


So what does this look like? 

How do we integrate various sources of information in eLearning? 

What are some goal-free problems we can use to illustrate a point in eLearning? 

If repetition is the key to learning, how do we reduce redundancy? In other words, where is the dividing line between necessary repetition (in whatever form) for reinforcement and unnecessary redundancy? 

And let’s just ignore the statement of the obvious –“use visual and auditory instruction techniques…where both types of instruction are required” – since it is hardly more than a tautology, particularly when he doesn’t identify examples of good and bad application of the various stimuli.

So what does he provide as advice?

Here are some tips for how you can reduce cognitive overload in your eLearning course design:

  1. Keep it simple 
    Remove all content that isn’t absolutely necessary for the learning process. For example, if you are designing a slide show to provide information, try to reduce the amount of extraneous graphics you use throughout. 
  2. Use different instructional techniques
    Present information in different ways. For instance, offer some data verbally and other data visually, such as through images or graphs. This will allow the learner to absorb information using different processing methods, which will reduce cognitive overload. 
  3. Make learning “bite sized” 
    Divide content up into smaller lessons and encourage them to only move forward with the course when they have fully grasped the current material. This will insure that they do not overload their working memory and can effectively move the information to their long term memory. 

Puh-leez. I miss the low-key wisdom of my colleagues like Andrea Mitchell who not only did a presentation on cognitive load, but had far more helpful words of advice than this. Please indulge me while I deconstruct (contradict) what Chris says.


 Despite the fact that learner engagement and memory is enhanced by more complex graphics than simple ones, Chris advises designers to keep things simple. Note the lack of nuance or qualified treatment of how excess simplicity can actually impede learner retention. Sometimes, things need to be complex to be engaging. For instance, which pie chart is more engaging and memorable:






Well, that’s not entirely fair, but it does make a point. Which one do you want to look at longer and which one has more useful information that you are likely to recall later on?

Simplicity is not always good, just like complexity is not always bad or hard to remember. How many of you can remember how to pronounce that ridiculously long word from Mary Poppins? It was the complexity that made it memorable (and the repetition through  music), not simplicity.

Different Techniques

As referenced at the beginning, this is helpful advice that tells us nothing. Might as well tell us to reduce overhead and increase sales. Ya’ think?

How about we use APPROPRIATE instructional techniques? What a concept! Here are some recommendations from the real world:

  • When the performance objective is for learners to state a policy or describe a practice, use a verbal instructional method.
  • When the performance objective is to “Identify the parts on the following piece of equipment,” it is best to use either a visual guide of the piece of equipment or a hands-on treatment of the equipment in question.
  • When the performance objective is to “Identify the third movement of Beethoven’s Fifth Symphony,” it is best to provide an auditory instruction method so learners can hear when the movement begins and identify  transitional musical phrases.

Are you getting the picture? Some things are obvious. Maybe he should have said to match your instructional method to performance expectations. We don’t train students to drive cars by methods different from the way they will be tested. Why would we do workplace training any differently? Okay, there are reasons why we would do it differently relating to what senior management mandates, but you understand my point.

Bite size

Ripley’s Believe It Or Not was bite size and often very memorable. My kids frequently spout little factoids, so I know they are persistent in long term memory. However, Chris never defines what a “bite” is and he certainly doesn’t tell us what to do with the bites.

IDs know that contextualizing and providing a cognitive framework for organizing the information is far more important than bite sizing. Let me elaborate, particularly in the context of cognitive load.

Pappas is fond of citing Miller’s Law (Memory can hold 7+/- 2 bits of information) without having actually read Miller’s paper which says nothing of the sort. See the original paper for yourself: http://psychclassics.yorku.ca/Miller/  Point being, Miller proposed that there were ways to work through the span of immediate memory and he proposed a number of ways to do this.

Miller pointed out that his experiment was limited to unidimensional stimuli. Read what he says about recoding. Also read what he says about multidimensional stimuli. Here’s an illustration of the two different kinds of stimuli with relation to short-term memory, and presumably cognitive load.

  • The game “Simon” is based on the ability of people to remember a series of four tones played in random order. Most people fall in the 7 +/- 2 category (or less) for the number of random tones they can remember in the short term. I know of no one who can duplicate the pattern of tones from any game they’ve played in the long term. In other words, the learning is not persistent over time.
  • The song “Twinkle, Twinkle, Little Star” is based on five tones played as a phrased series of 42 musical notes that many people can memorize after hearing it just one time. Additionally, the song can be sung or played decades after hearing it the first time, demonstrating its persistence.  

Not only are there more notes played, but the duration of the notes also varies, and folks find it incredibly easy to duplicate this more complex string than the radom, uniformly timed notes of Simon. When it comes to music, it’s often easier to remember longer, more complex pieces than it is to recall short, random pieces. So, despite Miller’s Law, it is really difficult to find an upper limit to what humans can remember when it comes to certain stimuli. For instance, we can remember a startling number of faces and some individuals have phenomenal performance at this. See: https://www.sciencenews.org/article/familiar-faces


In summary, his three bits of advice are both unhelpful without a context or caveat and in some cases just downright wrong when presented as unqualified “rules” for instructional designers. 

However, in the interest of providing candles instead of just cursing the darkness, I refer you to this helpful little bit from Jon Matejcek: http://www.dashe.com/blog/elearning/improve-learner-retention-forgetting/  He offers a singular solution to the “forgetting problem” from Don Clark – Spaced Practice. In other words, a quick and easy solution to increasing learner retention is to build in “regular rehearsal and practice…over a period of time.” Instead of fixing short term memory as the eLearning blog suggest through clever tricks, instituting a performance support system that offers practice over time will be far more effective at long-term retention. I would recommend following the links in Jon’s blog post for more helpful treatments on the topic of forgetting.