Cognitive biases in UX design - what are they? 8 examples
In our previous article, “Psychology and UX Design” we looked at human-computer interaction (HCI) and the importance of taking human psychology and cognitive principles into account when designing the UI and UX of digital products. In this follow-up piece, we examine the different types of built-in bias that we all carry around with us and how they impact UI/UX design.
Table of contents
What is cognitive bias?
Cognitive bias occurs when we subconsciously use stereotypes and existing mental models to speed up the thinking process and decision-making. However, while it’s easy and natural to fall back on our biases, this kind of shortcut thinking often leads to errors of perception, misinterpreted information, and false conclusions. When it comes to UX research and product development, users, product owners, UX designers, development teams… everybody is prone to bias. The key is to design the user interface and overall experience so that the user’s goal in using the product is not derailed by their own biases. And for product owners and designers to not impose their own biases on the product.
Confirmation bias – we see what we expect/want to see
Confirmation bias is the tendency to look at objective facts selectively, favoring those that confirm your existing beliefs and subconsciously ignoring information that doesn’t align with your own perspective. The danger with confirmation bias is that we don’t validate our hypotheses and decisions with a sufficient amount of research and data.
Overcoming confirmation bias requires designers and developers to question their own decisions, embrace contrary opinions, and not to rely solely on either subjective opinions or just a single source of information. The more evidence you gather and consider, the less likely confirmation bias will skew the end results.
Optimism bias – we ignore the risks
Our optimism bias leads us to think that we are less likely to have negative experiences than others. When it comes to building a product, optimism bias can make us more likely to neglect apparent risks and just hope for the best. Over-confidence in our ideas can lead to underestimating the value of assumption validation and risk management.
The best solution is to do the opposite – actively look to uncover potential risks as early as possible. Having identified them, you are in the best position to mitigate them and minimize their impact on your product.
False consensus bias – we think we’re all the same
We generally have a tendency to believe that other people share our opinions and beliefs, and would behave the same way we would in a given situation. The UI of your product might seem perfectly clear to you, most likely because you’ve already spent time working on it. However, that doesn’t mean that your users are going to experience the same degree of clarity. People are all different, from different backgrounds, and with different experiences – what’s obvious for you might not necessarily be as intuitive for your users.
Overcome false consensus bias by testing, testing, and testing again. Find out the issues your users are facing, not the issues you think you would face.
Recency bias – sometimes, we have the memory of a goldfish
We focus on recent events – and we tend to remember what we’ve recently heard or experienced better than the more distant past.
Push back against recency bias by keeping notes and records. For example, if you’re interviewing users, there’s a lot of information coming your way and you need all of it. Look for ways to include a notetaker in interviews or other meetings, or get attendees’ permission to make a recording so you can be sure all the evidence is stored and easy to access and analyze.
Sunk cost fallacy – we don’t like to waste an investment
Once we’ve put a degree of time and effort into an endeavor, we don’t like to see it wasted. Not even when everything is telling us that the best option would be to stop or change course. This can be particularly true of development teams and product owners. If the business goals or user needs change, we are reluctant to pivot because of the distance we’ve already traveled - even when the objective data shows we’re heading in the wrong direction.
Avoid the sunk cost fallacy by investing in research and analytics, collecting data from day one. And for each element of the product, set up success metrics so you can definitively tell if it’s accomplishing the intended goal.
The IKEA effect – we built it ourselves, it must be good!
IKEA is famous for its ‘build-it-yourself’ offerings and when we have created a product, or participated in its creation, we tend to perceive it as being of higher value than it actually is.
How does this help? Well, by letting our users customize or partially create their own product experience, we create a bond between our product and the people who use it. You can use the IKEA effect to increase users’ emotional involvement, and the value they perceive in the product.
However, from the development team’s perspective, the IKEA effect can backfire, when we fall in love with the product ideas and features that we have created (in that sense, it has a similar effect to the sunk cost fallacy). Designers, developers and product owners must take the role of objective observers and researchers, using critical thinking, exploring different options, and running experiments.
Framing bias – it’s all in how you ask
When you ask a question, or present information, the response depends not only on what you’re asking or presenting, but also on how you’re asking or presenting. Leading questions result in distorted data. Effectively, your biased question gets you a biased answer. Neither of which are good for product development.
The most efficient remedy for framing bias is to ask broad non-leading questions, such as “How would you describe your experience with our website?”, “How would you rate the product?” or “Could you tell me what was difficult and what was easy while filling out the form?”. This way you’re much more likely to uncover some true insights instead of hearing answers by your question.
Social desirability bias – we like to fit in
Even if you do frame your questions neutrally (avoiding framing bias) people may still give less than honest answers. As humans, we like to please others, and also make ourselves look good – in other words, research participants might be inclined to give the answers they think you would expect from them, or that would put them in a better light. They might also give more positive feedback on a product or feature because they don’t want to offend the interviewer.
Social desirability bias is more likely to occur during in-person contact, e.g. a focus group or a user interview. To minimize the impact on your research results you can:
- Reassure the interviewee that they can be 100% honest, and that the information received will be absolutely confidential and only used to improve the product.
- Be as neutral as possible to avoid your own opinions guiding the user.
- Use open-ended questions as much as possible.
Cognitive bias and digital product development
When considering human-computer interaction and human psychology, cognitive bias is a significant and widespread factor (we all have them!) For UI/UX designers, cognitive bias forms part of how their intended product users operate… how they will perceive, engage with, and use (or misuse) the product. Understanding common assumptions and unconscious preconditions – both the users’ and your own – is critical to successful digital product design.
Authors:
Sylwia Rapacz - QA Engineer, a student of the Masters of Science in Computer Science at Georgia Tech, with courses like Human-Computer Interaction and Introduction to Cognitive Science
Kateryna Kaida - Product Designer at Boldare
Share this article:
