Glossary · 7 · Evaluation methods
Cognitive walkthrough
Also known as: Cognitive walk-through
German: kognitiver Walkthrough
In usability engineering, a cognitive walkthrough is an expert inspection method in which evaluators step through a task with a product, action by action, from the perspective of a new user who learns by exploring, and ask at each step whether that user would know what to do, find the right control, connect it with the goal and understand the feedback.
- Applied cognitive psychology
- Evaluation methods
In one sentence
A cognitive walkthrough is an expert review that steps through a task from a new user's view, asking at each step if they would know what to do.
Example
Before user tests, two writers walk through the procedure “Replace the label roll” on the HMI and in the manual. At step 4 they note that a new user would not know that the splice table is behind the hinged cover, and add a figure.
Rests on
- Gulfs of execution and evaluation: the walkthrough's questions ask whether the user can form the right intention, find the action and evaluate the result.
- Recognition and recall and affordance: exploring users act on what they see; labels and visible controls must suggest the right action.
- Mental model: whether the user's model of the task matches the product's.
How it applies
- Procedure reviews: Walk through each step of an instruction with the product or the HMI: Is the goal of the step clear? Is the control named as it appears? Does the reader know the step succeeded?
- HMI texts: Check messages, button labels and help texts at the point where a new user meets them.
- Before testing: A walkthrough is cheap and finds obvious problems before usability testing with users, which it does not replace.
- Software documentation: Walk through onboarding and installation guides as a first-time user.
Cognitive walkthrough vs. heuristic evaluation vs. usability testing
A cognitive walkthrough follows specific tasks from a learner's view. A heuristic evaluation checks a design against general principles without a fixed task. Usability testing observes real users. The first two are inspections by experts.
Limits
- Evaluators are not new users; they miss problems their own knowledge hides (curse of knowledge), and results depend on the chosen tasks and personas.
- The method focuses on learnability for new users, less on efficiency for experts.