Cognitive Load Assessment
Assess cognitive load across three dimensions and provide actionable recommendations to reduce unnecessary burden on users.
The Three Types of Cognitive Load
Intrinsic load — complexity inherent in the task itself. A tax return is intrinsically complex. You cannot simplify the tax code. Focus: break into manageable steps, provide context, explain terms.
Extraneous load — complexity added by the design, not the task. Confusing navigation, unclear labels, inconsistent patterns, visual clutter. Focus: eliminate ruthlessly. Every instance is a fixable design failure.
Germane load — effort invested in the actual task. This is the load you WANT. The user engaging with their real goal. Focus: maximise this by removing the other two.
Assessment Framework
For each screen or step, evaluate:
| Dimension | Low (good) | Medium (review) | High (redesign) |
|---|---|---|---|
| Decisions required | 0–1 | 2–3 | 4+ |
| Items to remember | None | 1–2 | 3+ |
| New concepts to learn | None | 1 | 2+ |
| Steps to complete | 1–2 | 3–5 | 6+ |
| Reading complexity | Plain language | Some jargon | Dense/technical |
| Visual complexity | Clean, focused | Moderate | Cluttered |
If any dimension rates High, that screen needs redesign before launch.
How to Assess
- Walk through the flow as a user who is tired, stressed, and unfamiliar
- At each screen, count: decisions, things to remember, new concepts
- Note where you pause, re-read, or feel uncertain — those are load spikes
- Map the load type: is the difficulty intrinsic or extraneous?
- Recommend specific reductions for each extraneous load source
Reduction Strategies
- Progressive disclosure: show only what the current step needs
- Smart defaults: pre-select the most common option
- Chunking: group into sets of 3–5 related items
- Recognition over recall: show options, don't make users remember
- Persistent context: show key selections throughout a flow
- One primary action per screen: de-emphasise secondary actions
Output Format
Present the assessment as a screen-by-screen table showing load ratings, followed by prioritised recommendations sorted by impact.