Ease of use is not ease of learning

Most products are designed to be effortless. Learning is the one domain where difficulty is often the point.

Remove friction. Shorten the path. Anticipate the need. Reward every action. Those instincts are right for shopping, streaming and social feeds. When the goal is learning, they mislead.

What feels best now often teaches least

Predict before you look

Students studied a prose passage. Then half re-read it, and half took a recall test on it with no answers shown.

A week later, which group remembers more?

Bjork and Bjork call this a desirable difficulty. Performance during practice is a poor guide to what has been learned. The two routinely pull apart. A team that measures only what happens inside the session is measuring the wrong curve.

Learners choose what feels effective, and they are usually wrong

Spacing beat cramming. People learned painting styles either one artist at a time or mixed up. Mixing taught the styles better. Asked which worked, most chose one-at-a-time, even after seeing their own results.
Active classes taught more. Harvard physics students taught the same material through active learning scored higher than those given a polished lecture. They also felt they had learned less, and rated the instructor lower.
Fluency flatters. A confident, fluent lecturer raised students’ sense of how much they had learned without raising how much they had learned.

Satisfaction, perceived ease and self-reported learning are real signals about experience. They are not signals about learning. Treat one as the other and you will optimise the wrong thing, with excellent survey scores.

A learner who feels fluent is not necessarily learning. A learner who is struggling productively often is.

That doesn’t mean learning products should be hard to use

Confusing navigation and cluttered screens waste effort that should go into learning. The discipline is telling two kinds of difficulty apart.

✓ Protect

Difficulty that is the thinking

Recalling rather than recognising
Working out a step before it’s shown
Meeting an idea in an unfamiliar case
Getting it wrong and finding out why

× Eliminate

Difficulty that is the interface

Hunting for the next action
Reading a label and its legend in two places
Decorative motion competing for attention
Holding instructions in mind while doing the task

Cognitive load theory names this. Intrinsic load comes from the material: manage it by sequencing and chunking. Extraneous load comes from the presentation: it is waste, so cut it. Germane load is the effort that builds understanding: leave room for it. Its own authors admit the three can’t yet be measured separately. Use it as a design heuristic, not a measuring instrument.

And a limit worth keeping: if a learner can’t meet a difficulty successfully, it isn’t desirable. It’s just difficult.

One question for every screen

Does this help the learner do the thinking that causes learning, or does it just make the experience feel good?

If a principle your team writes can’t be checked against this question, it is probably about the product rather than the learner.

Before you scroll up

Without looking: what are the two kinds of difficulty, and which one do you protect?

Take this with you

Performance in the session is a poor guide to learning. Measure later.
Learners prefer what feels fluent. Their preferences are real data about experience, not about learning.
Strip the difficulty that belongs to the interface. Protect the difficulty that belongs to the thinking.

Our job isn’t to make learning feel effortless. It is to make the right effort possible, visible and worthwhile.

Try this week

Mark the friction in one learning flow.

Walk through one flow in your product as a learner. Note every moment of effort.
Mark each one P (protect: it’s the thinking) or E (eliminate: it’s the interface).
Find one place where the product shows an answer, and redesign it to ask first.

To mull over

Which of your product’s success metrics would rise if it taught less?
Where has your team removed a difficulty that was actually doing the teaching?
How would you explain to a stakeholder that lower satisfaction might be good news?

Sources

Roediger, H. L., & Karpicke, J. D. (2006). Test-enhanced learning. Psychological Science, 17(3).
Bjork, E. L., & Bjork, R. A. (2011). Making things hard on yourself, but in a good way. Soderstrom, N. C., & Bjork, R. A. (2015). Learning versus performance. Perspectives on Psychological Science, 10(2).
Kornell, N., & Bjork, R. A. (2008). Learning concepts and categories. Psychological Science, 19(6).
Deslauriers, L., et al. (2019). Measuring actual learning versus feeling of learning. PNAS, 116(39). Carpenter, S. K., et al. (2013). Appearances can be deceiving. Psychonomic Bulletin & Review, 20.
Sweller, J., van Merriënboer, J. J. G., & Paas, F. (2019). Cognitive architecture and instructional design: 20 years later. Educational Psychology Review, 31.