Cognitive load (German: Kognitive Belastung) describes the amount of mental demand a task places on a user's working memory. Because working memory is a tightly limited resource, the likelihood of error rises sharply as soon as the demand reaches this limit, which makes it a central starting point for designing safety-critical user interfaces.
Working memory as a limited resource
Working memory holds information that is needed for the current action but is neither permanently stored nor currently visible. Its capacity is small: common estimates are in the range of a few units available at the same time, and without active rehearsal the content decays within seconds.
A directly applicable design rule follows from this: Every requirement to remember something is a load that should be avoided. A user who has to read a measured value from one screen, keep it in mind, and enter it again on another screen is working at the capacity limit, and an interruption at exactly that moment reliably erases the information. Such transfer steps are one of the most productive sources of error and at the same time one of the easiest to eliminate.
Three types of cognitive load
The relevant theory distinguishes three components. Separating them matters in practice because only one of them can really be addressed through design:
- Intrinsic load results from the complexity of the task itself. A ventilator setting is demanding in terms of content, regardless of how well the device is designed.
- Extraneous load arises in addition from the way information is presented: nested menus, inconsistent units, required conversions, scattered information. This component is purely a consequence of design and is therefore largely avoidable.
- Germane load is the effort that goes into building a sound mental model. It is not harmful but desirable, provided that capacity remains available for it.
The practical consequence is clear: The goal of design is not to minimize cognitive load in general, but to eliminate the extraneous part specifically. The capacity this frees up is available for the actual professional task and for building competence.
Task demand and individual strain
In German occupational science, two terms are distinguished that often merge in English: Belastung (task demand or load) refers to the demand acting objectively from outside, and Beanspruchung (strain) refers to its individual effect on the specific person. The same load leads to different strain in different users, depending on experience, condition on the day, routine, and concurrent load.
For study planning, this distinction is more than terminology. It explains why an evaluation based on an experienced, well-rested test participant is systematically too favorable, and why the user profile must represent the range of the user group rather than its average.
Drivers of high load in medical technology
- Memory demands across context switches: values, settings, or identifiers that have to be carried over from one screen, label, or device to the next.
- Conversions and inconsistent units: every mental calculation is an additional source of error under time pressure.
- Menu depth and scattered information: when decision-relevant details are not located where the decision is made.
- Interruptions: after every interruption, the state of the task must be reconstructed from memory; skipping a step on resumption is a typical consequential error.
- Parallel tasks and alarms: what counts is the sum of the demands at a workstation, not those of a single device.
- Stress and time pressure: under strain, the focus of attention narrows and the available working memory capacity drops further. In emergency use in particular, increased demand and reduced capacity therefore coincide.
How cognitive load can be assessed
Cognitive load cannot be observed directly, but it can be assessed with established methods, and this is the prerequisite for substantiating it reliably in an evaluation rather than merely asserting it:
- Subjective methods: standardized questionnaires such as the NASA Task Load Index (NASA-TLX) capture the perceived strain across several dimensions and allow comparison between design variants.
- Performance-based methods: a secondary task performed in parallel indirectly shows how much capacity the primary task occupies.
- Physiological indicators: for example pupil response or heart rate variability, supplemented by eye-tracking data that serve as indirect indications of search behavior and attention and are usually interpreted in addition to other methods.
- Behavioral indicators from observation: hesitation, paging back, repeated checking, verbalized uncertainty in the think-aloud method.
Comparing variants is more informative than an absolute value: It is easier to show reliably whether a redesign reduces load than whether a particular load is "too high."
Design measures to reduce load
The most effective measures follow a common principle: recognition rather than recall. In concrete terms, this means displaying decision-relevant information where the decision is made; carrying values over automatically instead of having them entered again; offering sensible defaults; keeping units consistent; and leaving conversions to the device.
Special attention is due to resuming after interruptions. A device that keeps the processing state visible and clearly shows, after a pause, where in the sequence the user currently stands, provides relief in exactly the situation in which most omission errors occur. Further helpful measures are a flat menu structure, reducing the information shown at the same time to what is relevant for the decision, and labeling that consistently follows the terminology of the target group.
Regulatory reference
Cognitive load is not a separately defined or required term in the medical device standards. It acts indirectly: use errors that result from overloading working memory must be considered in the usability engineering process according to IEC 62366-1 and minimized through design, in practice within the use-related risk analysis. For critical tasks, reducing extraneous load is particularly important, because even small losses of information can have safety-relevant consequences there. IEC 62366-2 addresses cognitive factors in its application guidance.
For psychological workload at the workplace, the DIN EN ISO 10075 series provides a separate ergonomic standards basis that, in particular, systematically separates the concepts of load and strain. The FDA Human Factors Guidance treats cognitive demands as a use-related risk factor that must be considered in the analysis and in human factors validation testing.
Working memory is a scarce resource, and every avoidable memory demand uses some of it up. The goal of design is not to lower load in general but to remove the extraneous load, the load that arises solely from the way information is presented.
Cognitive load is measurable. Those who assess it with established methods instead of estimating it can substantiate the effect of a design decision rather than merely asserting it.
Frequently asked questions (FAQ)
Is cognitive load a normatively defined term?
Not in the medical device standards. There it acts indirectly, through the evaluation of use errors whose causes lie in perception and comprehension. In occupational science, the DIN EN ISO 10075 series provides a separate standards basis for mental workload that distinguishes between load and strain.
What is the difference between load and strain?
Load (German: Belastung) is the demand acting objectively, strain (German: Beanspruchung) is its individual effect on the specific person. The same load strains an experienced, well-rested user differently than an inexperienced user at the end of a night shift, which is one reason why study samples must represent the range of the user group.
How can cognitive load be measured?
Through standardized questionnaires such as NASA-TLX, through secondary-task methods, through physiological indicators such as pupil response or heart rate variability, and through behavioral indicators from observation. Comparing design variants is more informative than an absolute value.
Which design measure works fastest?
Eliminating memory demands. Wherever a user has to carry a value from one place to another in their head, load can be reduced immediately through automatic transfer or simultaneous display, with a direct effect on the likelihood of error, especially in the case of interruptions.
Is high cognitive load always bad?
No. The intrinsic load of a professionally demanding task is often unavoidable and not a problem in itself. Extraneous load is the critical part: it arises exclusively from the design of the system and occupies working memory capacity without any professional added value.
Do you want to measure the strain on your users rather than estimate it? We combine measurement data, expert assessment, and medical interpretation into an objective analysis.
More about our ergonomic analysisSources
- IEC 62366-1:2015+AMD1:2020, Medical devices, Part 1: Application of usability engineering to medical devices
- IEC/TR 62366-2:2016, Medical devices, Part 2: Guidance on the application of usability engineering to medical devices
- FDA Guidance: Applying Human Factors and Usability Engineering to Medical Devices