Cognitive load and cognitive biases

Cognition refers to the various processes that occur within our brain. Though user research sessions are not psychological studies, you still need to consider that people think, interact with, and interpret information in many different ways.

Your participants will all have different levels of ability and capacity. When designing and conducting user research, there are two key concepts you must consider:

Cognitive load

Cognitive load is the mental effort required to process information and complete a task. The more complex, unfamiliar, or distracting something is, the harder it becomes to think clearly.

Cognitive bias

Cognitive biases are unconscious, automatic processes in our brains that are designed to make decision-making quicker and more efficient.

Biases help us to manage information – they are not inherently negative, but they also can lead to negative assumptions and poor decisions.

Cognitive load in user research

There are three types of cognitive load – intrinsic, extraneous, and germane.

Intrinsic load refers to the inherent complexity or difficulty of the task itself, without any external influence.

In simple terms, an easy or simple task has a low intrinsic load, whereas a more complicated or difficult task has a high intrinsic load.

Extraneous load refers to the way tasks are presented to the user.

Simply put, it means the way things are designed or how instructions are provided. Complicated or unclear instructions increase the extraneous cognitive load by making it more difficult to understand what the user is actually being asked to do. Extraneous load can also be increased by a noisy or distracting environment.

Germane load is the effort needed to learn something new – even easy tasks take additional effort to commit to memory. Germane load is affected by the levels of intrinsic and extraneous load, as these affect how much effort the user is able to dedicate to learning and remembering, rather than trying to just work out how to do a task or deal with external stresses and distractions.

Minimising cognitive load

It is important for you to be aware of these when designing your user research, as your participants will only be able to tolerate certain levels of cognitive load. The threshold will vary depending on the specific task and who your participants are.

This is where your awareness of accessibility and inclusion is important – everyone uses their brain differently and will have different responses to cognitive load.

It should be obvious that there will be very few people who would be able to maintain focus throughout an intense, 3-hour long interview session followed by several complex problem-solving tasks, so this would be very poor user research design.

However, although this is an extreme example, you need to bear in mind that even a 30-minute session might feel like 3 hours for some participants, or that a task you think is simple can appear very complex to someone else.

When planning your user research, think about how you set up your sessions. Keep test scenarios simple with clear instructions.

Begin with straightforward questions using familiar words, and avoid abstract notions or concepts. This helps everyone and especially those who find it difficult to process a lot of information or instructions in a short time. If you overwhelm your participants they may reach their limits of cognitive load and no longer be able to participate effectively in the session. This can affect the quality of your results – when people become tired, they cannot properly participate in cognitively taxing tasks.

Be aware of energy levels. Fatigue can sometimes be an issue for some participants with disabilities due to the condition they have, the medication(s) they take, or the extra effort required to use assistive technology. During the recruitment process, ask participants about any time and fatigue considerations, and if there are any times of the day they can focus on tasks more easily. Then plan session timings accordingly.

Even if you think a task is easy or that you have scheduled enough breaks, you need to be ready to adapt to accommodate the needs of all your participants.

Cognitive biases in user research

In Week 2, we learned about bias as a form of risk in user research that can potentially lead to skewed results. This week, we will explore this in more detail, as there are multiple different types of cognitive bias.

Here are some that are particularly relevant for user research.

Confirmation bias

Confirmation bias is the tendency for people to search for or interpret information in a way that confirms what they already know or think. They look for what they expect (or want) to see while discrediting or ignoring information that conflicts with their initial opinion or expectation.

This occurs much more frequently when people are closer to something emotionally.

For example, let’s consider peoples’ attitudes towards their favourite singers or sports team.

People are often more likely to agree with and promote things that make them look good – such as sold out concerts or winning matches – but they may then discount exceptions to this, or even ignore negative press regarding controversial actions or opinions the stars may have. This is because they want to maintain their positive view of their favourite celebrity, so they look for ways to confirm it.

In user research, confirmation bias can occur when you want to prove a hypothesis that you believe to be true. For example, if you think a user will struggle when trying to phone a government department to get answers to their question, you could be more likely to look for evidence that proves that the user is struggling rather than the evidence that they are not. Alternatively, you may be inclined to stop gathering evidence, because you have retrieved the evidence that confirms your hypothesis already.

This will lead to you missing insights.

Anchoring bias

Anchoring bias occurs when our judgements or decisions are influenced by a reference point (or “anchor”), which is often totally irrelevant.

When we make decisions, we often try to rationalise or justify them based on what we already know. In user research sessions, this could be information that participants already know, but it could also be information introduced early in the session that influences later decisions.

Anchoring bias also refers to peoples’ inability to make adjustments to their assumptions even when given the “correct” answer – they try to rationalise their logic based on what they already know, rather than accepting the ‘new’ information.

In user research, anchoring bias can occur when a participant focuses on the first topic of conversation or question in a survey, and uses this to ‘frame’ the rest of it. It could also occur during usability or A/B testing. There can be a tendency to prefer the first version of something the user sees, even if the first version is more complicated, a user may tend to prefer it simply because they saw it and figured it out first. Another example is default or ‘suggested’ entries in forms.

Status quo bias

The status quo is the way things currently are. Status quo bias means the preference that people often have for keeping things that way.

As with other forms of bias, this is not necessarily a bad thing. Not everyone likes to take risks, and innovation just for the sake of innovating is not always the best idea.

However, change is often a good thing. Processes can be made easier, simpler, and more streamlined, which benefits everyone. But sometimes people need a gentle nudge towards trying something new, rather than just doing things the way they have always done it.

The transition to digital services is a big step for some people, and as technology advances even further, the gap between those who prefer not to use or who are unable to use certain forms of technology will get wider.

When we design government services, we have to carefully balance the genuine needs that some people have for different forms of access to services, and those who might simply resist change.

Social desirability bias

Social desirability bias occurs when people provide answers that they think will be viewed favourably, rather than answering honestly about their true beliefs, opinions or experiences.

People want to fit in, make friends, and create bonds with other people, and they do not want to be viewed as problematic or undesirable. This means that research participants will sometimes tell you what they think you want to hear, rather than the truth, which can distort your findings.

This can occur when researching sensitive topics – people may want to seem to be socially conscious or ‘politically correct’ rather than giving their true opinions. Some participants may also want to hide behaviour that they fear may have negative consequences for them, such as discussing drug use.

It can be difficult to identify when this occurs or account for it in your findings, but you should always make clear to your participants that research sessions are safe spaces where they can answer honestly and their data will be kept confidential. This can help make them feel more confident about speaking openly.

Everyone has biases. Bias is not necessarily always a bad thing, but you do need to be aware of it, and always to mitigate any negative influence bias may have. It is equally as important for you as a user researcher to mitigate your own biases as well as to design user research that mitigates the biases of your participants.

Task

Were you aware of these different types of bias and the effect they can have on user research? Can you think of other types of bias, or different ways that they could affect research outcomes that we have not discussed?

Share and discuss your thoughts with your fellow learners in the comments below.