That Never Even Crossed My Mind: How to Spot and Fix Cognitive Biases

Transcript

hi and welcome today's talk is going to be about cognitive biases and how to spot them and overcome them but will not be about dark patterns and how to exploit the biases of your users so the top will cover how the human brain works how the decision-making process is working how cognitive biases are affecting researchers the development team the

stakeholders the most encountered cognitive biases in user research and a way to overcome them so how does the human brain work we think about ourselves as rational human beings but usually the decision making process is actually working outside of our awareness so in his book Daniel Kahneman in his book Thinking Fast and Slow he defines the decision-making process in two systems system 1 which is very fast with effortless its intuitive and it's the one that helps us run out of danger

so if you would see a bear it's like snap judgment you just run system 2 is very slow and it's actionable when our attention is drawn to a problem so that's when system system 2 is taking place and it requires a lot of attention so both of these systems are impacting our reasoning but the primary one is system 1 which is very fast okay okay so Daniel Kahneman in 1974 has

defined cognitive biases as a systematic error in thinking that impacts one's choices and judgments which means we lose our reasoning and logic in order to be more fast this is like our brain takes shortcuts in order to be very very fast so from 1974 the researchers have mapped 194 cognitive biases but in user research we are going to talk about only 6 don't worry so we actually cognitive biases are impacting user researchers

designers stakeholders development team we are all biased we can't avoid that we all make the same mistakes but it's very important in user research because it's a continuous fight against bias to know what are them and how to see them so usually if a bias is it's very in-your-face like if you're conducting user testing and you are making leading questions like did you like this didn't you how about this awesome feature these are leading questions and are very easily to correct the problem is when

you the bias is hidden and you cannot see it exactly super fast so this is when you are collecting all of your data all of your data is pristine because you had a great script for the interview but you are biasing it while analyzing it and interpreting it so this is this is very I mean this is harder to correct it so let's see the most encountered cognitive biases in user research so confirmation bias this is when you are giving more value to

answers which are confirming your assumptions so this happened to me I'm from Alessio I am a designer and while I was doing user research for our block block explorer we we had 5 interviewees to whom we have shown the block overview page and only one of them got exactly what it was about and even though the other four didn't understand anything the development team came all of the time no but he understood so this is the confirmation bias we have to know about it so the hack for this confirmation bias

okay so hack is to assign someone on your team to play the devil's advocate for major decisions or use the six thinking hats technique which is thinking from different angles okay anchoring bias this is usually very much used in business negotiation so if I was to pitch you a project I would go like this this is a web app usually they use Google material icons but for this web app especially we have hired the

designer and these icons are unique and only made for this app so now you're in your mind you have this term of uniqueness and even though you didn't like the color theme or you didn't like it at all of it first you'll like it now because you have this uniqueness in in your mind and you're anchoring everything to that information so this usually happens when you are user testing and you're testing flows the user might say that he liked the very first flow because he got he understood that one so he will say that nevertheless so in order to to overcome that it's better to reflect on your

decision making history and think about whether you've rushed to judgment in the past so you loop into a system - so - to avoid that it's very easy to swap every time you're doing user testing swap the flows first this one then B or a bit as framing effect that is the way you are framing the information so it's very there has been an experiment in which people were described two types of vacation the first description was very bland

nothing extraordinary the second one had very like it had much superlatives so they would describe it brilliant beaches lovely sunrises awful nightlife and awful accommodation so because and after that the experimenters asked which one would you cancel and when they they asked which one would they cancel people were 94% even for the first one but for the second one

they wouldn't cancel it because of those superlatives so this is this is the framing effect so how to hack it is be aware of every words nuances and the way they can affect your interview participant so everything you're saying it might bias your interview observe very expectancy effect this this is a very tricky bias so if I have a bad day or if I am told okay you are going to test with power users these guys are brilliant developers they know exactly

what they say and I go there I have in my mind a result already I am expecting someone something from them and if they are not acting as I was expecting it or if there are not power users I'll have the tendency to be more arrogant or I don't know my body language will somehow affect their answers so in the end I'll get some super biased results so in order to get rid of that is very important to screen yourself and your team for the potential cognitive biases

social desirability bias well this one happens all the time in every interview you are conducting so usually interviewees are thinking oh this person has super worked for this I want to make it feel good so they will not give you this year an honest opinion about your product and they will say something that they think that it will please you so in order to get rid of that tell the participants from the start I didn't work on this project I am just hired to facilitate this interview so so you make

them feel better and somehow get their honest opinion that you actually want curse of knowledge effect this is when you do not adapt your language to your users this usually I mean I have seen this many times in our eco space in etherium most of the times this is happening we don't tend to use common language too when we are talking to users so in order to do that and get their actual opinion we have to adapt our vocabulary use

common language with everyone and make sure that they understand you so how to fix and overcome our cognitive biases in user research so screen your study participants is very important to have the actually targeted audience for your interview not anyone from the street they will give you answers but nothing will happen list your assumptions determine the process of analyzing the results so this means you start that process you list it and you won't change it during the

experiment because this will be a total fail have a large enough sample size actually Nielsen Norman has a very big paper on this easier for so if you don't want to read it it's usually 5 from 5 the learning curve decreases and you don't get like tons after map your cognitive biases but I do this all the time I start and I map my cognitive biases because I usually am affected by hidden side effect which is like after I get the information I'm

thinking oh I knew that all the time I knew it further from the first time so do that do that with your team and do that with your potential participants write a script for interview and check not to have leading questions this is very important map your emotions so actually you cannot control yourself in a day if you are moody or you have a really bad day you can't do anything about it but if you have that bad day after you have harvested the information and you are writing down everything write that down the fact that you had a very bad day so you might go back and see that you might

have biased some stuff talk less and watch your body language very important and to recommended books for you Thinking Fast and Slow is brilliant and it has a lot of insights and hooked which is a book that helps you make your product desirable and make users develop new behaviors so I recommend you this one and this is me so any questions please okay no questions

yeah yeah I mean it was that time when when I was user testing and all my colleagues the development team because they usually come to that we're telling me no but he said that look he said that or when they were observers in the room because I usually do that via zoom and I invite them to see it because seeing is believing not just telling them I saw that look they said we're like I I can't understand how he doesn't see the button what's wrong so yeah this is any others thank you