The 4-Slot Desk: Why 'Just One More Example' Can Overload a Novice

You put up the prepaid insurance adjustment, then add "one more quick example" with an accrual thrown in. When you look up, the hall of faces looks like a trial balance that won't balance. So how much can a novice's head hold at once?

AMEE Guide No. 86 describes working memory (WM) as the bottleneck for learning. WM is the small, short-lived mental workspace that processes new information. Citing earlier research, the guide says it can hold about seven elements but actively process only two to four at a time.

Try this first: pick your hardest slide and list every new element a student must hold at once. If it's more than four, split it. That's our suggestion, drawn from the guide's principles, and untested.

What they found

WM also loses almost everything within 30 seconds unless it is rehearsed, which means repeated to yourself. An "element" is a unit of information. What counts as one depends on the learner, because several can combine into one chunk.

Long-term memory is a different story, because the guide describes it as theoretically limitless. It stores schemata. A schema is an organised chunk of knowledge that WM can handle as a single element, which is why expertise matters so much.

Cognitive load is the demand a task places on WM, and the authors define three kinds. Intrinsic load comes from the task itself, and extraneous load comes from anything not essential to it. Germane load is the effort of actually learning.

Because intrinsic and extraneous load add up, a task that sits comfortably inside an advanced learner's capacity can overwhelm a novice. That leaves no room for learning.

So the guide's toolkit works in three moves. First, cut extraneous load with techniques such as worked examples or putting text next to the diagram it explains. A worked example shows the solution steps before students try similar problems.

Next, adjust intrinsic load to the learner by sequencing from simple to complex. Then point the freed capacity at learning.

The guide also warns of the expertise-reversal effect. It means that supports that help novices may not help, and can even hinder, more advanced learners.

Before you rewrite every slide, though, it helps to know what kind of document this is.

What they did

John Q. Young, Jeroen van Merriënboer, Steve Durning and Olle ten Cate wrote AMEE Guide No. 86 for Medical Teacher. Rather than reporting a new study, it summarises cognitive load theory (CLT), which explains how WM's limits shape learning.

The summary includes CLT's model of memory and the techniques that CLT researchers have tested.

So how far can you push it?

The reality check: this is a narrative guide written for medical education, not a systematic review or an experiment, and it reports no new data.

A narrative guide summarises research the authors have chosen, rather than systematically searching for every relevant study.

The authors say most of the techniques have been tested in multiple studies. Even so, they also say CLT's application to medical education is relatively new and that valid ways to measure cognitive load are still needed.

They add that germane load's status as a separate category is debated. And nothing here was tested in accounting.