Evelyn & Kenny - How cognitive bias and ranking can foster an ineffective architecture and design

[00:00:15] Right. Great. Well, welcome everyone. And uh yeah, my name is Kenny and this is Evelyn. And uh we were here last year, well, we were here physically. Uh, but today we're going to talk about cognitive bias and ranking and fostering an effective architecture. So we always like to bring you back to a sort of personal story outside IT or outside of businesses. And um this this story goes a little bit more personal. It's uh my previous relationship which ended uh more than six years ago, I think. I'm happily married now. The reason I'm telling you it is is back then, it wasn't going so well. She was out alone on holiday and uh well, it wasn't really going good. So I was thinking, okay, what should what should we actually do to fix this? Well, as you know, I'm a software developer my background, so this is my flow chart of solving problems.
[00:01:14] Now, um that didn't went so well. We eventually also break broke up. So what I actually did was thinking, well, I'm a I'm a I need to fix this. I need to make this better. So in my mind, I was thinking, okay, what can we do? Well, let's have a fun holiday season, fun Christmas season. And that's what I tried to do and um well, that actually didn't work, right? Because we were dealing with complexity here. And when things go bad in this case, my own personal relationship, you need to endure, you need to observe what you can change and and probe that and see what can we change to make this better. And not go to a chasing a happy to be situation. And it's it's similar to like when your relationship goes bad, well, let's have a child or let's get married because then everything gets better, right? No, we need to endure and that's the that's the complexity. All right, but how does this relate to uh yeah, software? What do we do in software when things go bad?
[00:02:14] Well, pretty much uh the same thing, unfortunately. Uh, so in in software, at least what we are seeing uh is that we are also creating these happy to be situation in the complex environments that we are working on. And um we're seeing a lot. Everything that's presented here on the slides will probably ring a bell. It will seem familiar. Uh, we call this the copy paste hell. Uh, and as Kenny put it last week, he said, well, and it's mutating and that's actually what's going on. Um, but it's not what we should be doing probably. This whole copy pasting thing is not what will help us. And the reason that it won't help us is uh is indeed visible in on this slide. So, the the approach of copy paste doesn't work because we live and we work in these very complex environments. And what we tend to do is uh we tend to deal with complexity in a suboptimal manner because what we do is we treat this complex environment, which we are in, as complicated. Um and we treat them as as places that we can very easily change uh and and quickly change and that is not what these uh what these environments are about. So, this is probably not the best way to deal with them. And um that is why we strongly believe that we are dealing with complex adaptive social technical systems. We tend to think of our environments as as complicated and technical, but we are actually we have to balance these these three aspects of socio-technical system. And of course, in a social technical system, you have a social part, you have a technical part and you also have a cognitive part. And we will we will discuss that later. So you have to find a way to balance everything there because when you when you are working in a complex environment, you cannot just solve a gap. You really need to you need to listen to the system. You need to be able to inspect and adapt and probe, you need to try out stuff. You cannot just copy and paste. You really have to listen to your system and make really small decisions. Um, and when we take a look at a very real life example that is probably, um, yeah, familiar for um for every one of us, the open office space. I mean, uh I cannot see anyone raising hands, but usually when we ask this question, who works on an open office space, we see a lot of hands going up. Um, so for example, here the architecture of our office spaces in normal life because now we work remote, we work online, so it doesn't it doesn't hold. But this is also an architecture, right? It's not a software architecture, but the way that we design our uh uh our work environments, our offices. And technically this is a very easy thing to do. I mean, we can just create this. But there are other aspects here at play as well. So we cannot just copy paste something because it worked for Spotify or Google or whoever. Um, we have to take into account that this also impacts how we work as human beings and how it affect our culture, so the social environment. And if we don't balance that, then we will only get negative consequences of choices like this because it actually there is research that that shows that this kind of workplace is actually really bad for our health. We are not happier um and it's it's actually the most expensive workplace environment that you can have. So if this doesn't fit your culture, then you shouldn't be doing it. So um we shouldn't just jump on that bandwagon and we do this in software architecture as well, right, Kenny?
[00:05:22] Yeah. So it's it's not always bad to do the open office space, but well, inspect and adapt, look at your context. Is it useful for the outcome of we have? And like Evelyn already said, it's if you design a monolith or part of a software landscape without boundaries or the wrong boundaries, you just architected an open office space for software teams. What does that mean is, well, look, if if you're there in your ivory tower, we call it ivory tower architects and you're thinking how the system should be, right? Because that's uh that's what we do in design and architecture and find the decisions that make that uh landscape. And you don't talk to your system, to your social technical system and you don't observe the system or listen to that system, you can might as well just created that open office space for software teams. And and that's something interesting that Kevin Henley talked about in the past and that's called habitability, which makes a place livable, like home. This is what we want in software as well, right? That developers feel at home and can place their hands on any item without having to think deeply about where it is. You know, if you're in open office space, it's constant chaos in your head because that's why we keep these headphones on that I have on now. Uh, because we want to enclose, we want to make it habitable, right? I'm not at home. I'm not I'm only sitting with my wife and my cats now. I I cannot stand if I have 50 people in my room all day, every day that and I'm an extrovert. So we don't want that. And you know, we're dealing with software, but even in if if you see it physically in hardware, that habitability is even happening incorrectly, right? As you can see the fireplace here or the the the the bicycle lane, right? It's already happening there. So you might as well see it will happen in our architecture. And what I what we want to say here is architectural design is system design and it's contextual design. So listen to your systems, to your social technical system and it's a brilliant quote by Ruth Malan, right? So what's in, what's out, what spends and what move between. We need to have these boundaries. And we are a big fan of uh an approach called domain driven design that essentially the core pattern there is the bounded context and bounded context is not a pattern that goes into system boundaries, this goes into linguistic boundaries. So where are our linguistic boundaries? If we look at the world, they're much more clear, right? And that's what we want for a software design as well. And to create these, we need to listen to the system and work with the system. Now, what happens when um we build a system in this case that doesn't match our social technical system is that that is at play. Then you get all these weird things. And well, recently uh during the lockdown, we had Belgium on one side and still I think is on total lockdown and the Netherlands isn't. So here you see a shop that stands on the border between Belgium and the Netherlands. And for us this represents that software architecture, that software application, in this case the shop called Seaman, that represents a software architecture that is in odds with this contextual with this organizational design. Because what happens in the shop, you see these boundaries popping up, well, it's horrible when you go into the shop coming from the Netherlands and want to buy some baby clothing. Sorry, not going to happen. Because what happens in the end is and this is Conway's law explained a lot easier by Ruth, if the architecture of the system and the architecture of the organization are in odds, the architecture of the organization always wins. Right? So what we should be doing is not be separated from each other and do everything um separated, so in our own context and in our own thing, we should actually help each other and collaborate on this, right?
[00:09:25] Yeah, and that is why we are a huge fan of collaborative modeling. But the the goal there and what we want to achieve is creating that shared sense of reality. And in the previous slide, I I like I really like that quote. It also relates to Conway's law, of course, maybe that's familiar to uh to most of us here. But the goal is to create that shared sense of reality, to be on the same page about what we are going to do, where we are at this point. trying to speak the same language and uh have the same starting point and the same perspective. That's really crucial. And one way to get there is uh by collaborative modeling. And we are uh huge fans of that. We cannot do that physically at this point anymore, unfortunately, but we have found ways to do it remote. Um, the visualization here is a very important factor in in learning and understanding each other. So, um, it's a practice of using requirement analysis and modeling techniques to create a shared understanding. It's really about coming together and do this collectively. So especially with agile architectures, we really need this collaborative modeling. Because, yes, it's easier to decide things in isolation. Sometimes it's sometimes it's way easier to just be in your own bubble and do things. Um, and one of the reasons why that is easier is because it means that we can avoid or we can suppress some social challenges that are going on. We don't have to collaborate with anyone, so that means we don't have to have these social interactions. But that won't leave us with the best decisions. So we have to find a way of dealing with these social interactions and and stuff that's going there. So we have to have the right people in the room and then we need to take uh to take care of that interaction part properly. And uh one next question could be, well, who should be doing that? Um and well, we think that it should be a socio-technical architect, not a technical architect, but a socio-technical architect. And um this should be the one leading this. So this is not a role per se, it could be a role. Uh this is a role, sorry, but not a function. Um because if if we call it a function, then we also foster ranking. And we will get back to that later, but someone does need to own this. Someone does need to take ownership of this. Because otherwise, well, we, yeah, we won't be uh uh be very good at this. But a socio-technical architect is really what's missing at this point.
[00:11:32] Right, Kenny? Yeah. And but that sounds easier than it is, right? Just put a bunch of people in one room and think that they're going to collaborate and think that you will use all the knowledge in the room available and you come can come up with this architecture. So today, what we're going to discuss and I hope you now get the sense and the feeling that, well, we need to do more of this collaborative modeling. Um, and there are already was a nice question I see in the chat, so how do you know a decision is best? Well, we're going to uh tackle all these four subjects. So the first one we'll go into how to make sure everyone said what has to be said. Second, how can we create and include new insights? Third, who decides on the architecture and how to get everyone on board on the decision and who to divide and how diverse should people be? So define best, that's what we're going to do because that's uh
[00:12:18] There's not a real truth to that.
[00:12:20] No, that's our understanding, right? That's our perspective on that. let's be clear on that. Yeah. So the first thing is to ensure flow in meeting is how to make sure everyone said what has to be said. So I'm not sure if you ever done this before, recently, I was in a really nice uh workshop by uh Rebecca and Ken Power, Rebecca Rushworth and Ken Power. So the first thing once you are in a uh workshop, collaborative modeling workshop, see if you can ask the following questions to your participants. And that is how much were you able to contribute? Did you feel part of the discussion? How balanced were the contributions of the group? Did you feel heard within the group? Were there enough people in the room to make a decision? So this is what we call sense-making. And it's really nice to to start seeing patterns. So the the the nice chaotic dots that are there. That were the participants saying, okay, where on the scale are you? So left is none, right is a lot. And it's really nice to already start seeing what's going on in the group and start sensing that behavior. Because what you're essentially seeing here is the first part of our talk, so and that goes into ranking. Yeah.
[00:13:22] That's where goes in. Yeah. talking about ranking, right? Um, so this is uh this is indeed the first part and it relates to the questions that we that we asked you before. We we are doing this now in in our own workshops as well. Before we go into anything about ranking, just a sense making tool indeed to see where we are, how participants are feeling and then we are going to talk about something called ranking. Um, so there are um, well, ranking is a is is something that doesn't have to be a bad thing, but it's something that's going on all the time and we have a a difference between explicit ranking and implicit ranking. An explicit ranking is uh for example, your position in the organizational chart, your job title, your job description, your level of power, so whether you are a manager or or a CEO or whatever. Uh, it's also the the reason why Kenny could start with the micro story today. We always have a discussion about that with the small story about his personal life. So who could start? Well, he's the white tall guy in this session, so he is higher in ranking than I am. So um he could start. Um and we also have the implicit ranking and that's more uh stuff that is maybe a little bit more on the background or that you are less aware of. But it's very much influencing how you uh rank others compared to yourself. So your skin color, your level of charisma, how you communicate, uh the level of informal power for example, this sort of stuff is uh is implicit ranking. And like I said, it doesn't have to be a bad thing because it also helps us making sense of the world. But there are also some downsides of the ranking part and this is um how we are conditioned a little bit um because we do it unconsciously. Ranking is something that we are doing not very much aware. So it's also related to symbolic violence. I thought that Romeo was in the chat so I'm a bit scared now because he is the expert there and we are stealing his ideas, but he will be giving a talk about this tomorrow, so please go see that if you speak French. Um but before we go there, some examples of this ranking part. Um and for example, the one on the left, I really really like that one. Um I often see women asking men to treat them the same way that they the men treat other men at work. Um that has a lot to do with ranking and especially the tweet that's beneath that um about ignoring a power dynamic when you're on the upside, the upward side, uh you are reinforcing that dynamic. And this is what happens a lot in uh in our daily lives. So we have to be aware of this ranking going on. And the tweet on the right side of this slide, this is something that I could have written. Um because my story relates to that. I won at one point in my previous job. I was in a meeting and we walked into a room. I was the only woman and everyone started to introduce himself and uh they just said their names. And then they came to me and it was like, oh, how nice that marketing was involved. And I was like, I couldn't say anything at that point because I was just, I was stunned. But this is what happens. And it made me feel bad, right? It made me feel bad. Um, but I am doing that to myself mainly and that is where symbolic violence comes in as well. Uh, apart from that, it was just a very shitty thing to say and that he never should have said that. Um, but it's a it's the advantage of having a background in social uh social sciences as well that I start to analyze and ask myself, okay, so why does this did this make me feel bad? Um, because well, what I did there, I ranked myself below that person who said that to me. And that has a lot to do with all the implicit things that are going on there and the explicit things, but that's why it made me feel bad and that has a lot to do with symbolic violence.
[00:16:18] So, So it's the same that you just ranked yourself lower than Romeo. Yeah, it's a beautiful example of symbolic violence. Yeah, it is and I'm going into symbolic violence and now I'm even scarier, so please Romeo be kind to us.
[00:16:28] Um, but this is uh, I really really like this concept and he really inspired me last year at uh at New Craft in Paris as well. And symbolic violence means that we are all conditioned to have this symbolic ideal of what is dominant and strong and what is dominated and and weak. And we are conditioned in a way. Um and it's not really a judgment, it's more how society is wired. And this symbolic ideal changes over time as well. For example, the the cars that we are in, uh signs of wealth like the the watch you're wearing, um if you're being from if you're born in a rich country, uh if you're an extrovert, there are all signs that we think of, that we are conditioned to think of, as uh as being higher in uh in in power. So the more traits that you have of that list, the more power you yield to someone and the more power, yeah, the more the higher in rank that person is. So we all share that similar picture of this symbolic ideal. And we score people uh on that list and um yeah, the thing is that we are doing that to ourselves because we are scoring ourselves and others and we do that unconsciously. But the higher the score, the more power we yield to that person. And that is symbolic violence. So this is not done by this dominant person that we rank higher or score higher than than ourselves, but we do that to ourselves. So this is exactly what happened to me in that marketing meeting as well. And the fact that uh I wasn't aware of this dynamic going on then, made me feel very bad about myself. Um, I was kind of uh, well, they call it uh, well, Romeo, you call it hitting yourself. So I am um ranking myself lower, I'm starting to doubt myself, thinking that my my opinion or whatever I have to say is less valuable than that person that I rank higher. So, uh if we met that to a familiar situation and I'm going to quote you, I I already said that on uh on Twitter today. A pair programming or mob session where one of the participants is hitting oneself, uh like I explained just yet, is not a co-creation, it is submission. So, um, when you are pair programming with someone that you score higher than yourself, you are kind of conditioned to value that opinion over yours, for example. So you think that whatever you have to say or think is less valuable and you are uh you you are uh scared maybe to share stuff. So you you are tend to um not share stuff, you suppress knowledge at that point and that could be very harmful for that situation because you were doing pair programming and you want to contribute and you want to share that knowledge. This is also where the imposter syndrome comes in, of course. Um, but being aware of this rank and the score that we give to others and the score that we give to ourselves is really really important at this point to get the most out of our collaborations.
[00:20:23] Well, the next slide is how how can you battle with that, right? How should you start bringing stuff awareness? There's a nice mind game I saw Sasha today sharing on Twitter which I've seen before is just try to switch things around. So whenever you hear the word female or black person or whatever, right, a a connotation to that ranking, see if you switch it around. And if you find it odd, like male engineer, that's weird. Ah, wait, there's something there, there's a ranking going on there. And that's really nice, so turn it around mind game is one of my favorite. There was once this um article that says, I'm not sure who the who the who the person was was a famous male person, maybe George Clooney, that says uh why uh wife of George Clooney is making her name for herself. Huh? But why aren't you sitting putting her name in there, right? So there's also this this this thing what you can do is, okay, first switch it around. But the second thing what we try to do when we see that a lot of stuff ranking is going on there is try to do a ranking bingo. And a ranking bingo goes a bit. We we switched it one point to to French because we like to make that joke. But you can do a ranking bingo with each other and see how high your ranking opposed to the other, then stand in a line or make a virtual line that we do and then see if the the outer people can talk to each other, right?
[00:21:48] And ask them the question, okay, um, what is a situation when you were not being heard or when you were suppressing yourself, right? The symbolic violence again. This can start bringing awareness in then that also requires us to meet each other from person to person. So, these are just few things you can start doing in these collaborative modeling session if you see that's a lot of ranking going on there, implicit ranking. Because you have if you have explicit ranking like the boss is in the room, well, maybe that's not such a such a smart thing to do inviting your boss to a safe environment. And we get to that point a little bit later, why that's not a smart idea. But um yeah, what's what's actually so bad about this? Why is ranking actually so? Well, it's good, right? We're doing it to keep ourselves alive. But there's also a bad side to this ranking.
[00:22:37] Yeah, there is a bad side to this ranking. Um yeah, obviously, um like the example I just gave as well, um it can make you feel very bad about yourself and um you can get into some sort of crazy vicious cycle with with doing this to yourself. But the the main thing is that um the risk there is is that we are suppressing knowledge because we might be afraid to speak up. And uh if we're afraid to speak up, then that means that not everyone will be heard in a session. So if we're doing collaborative modeling and there's a lot of ranking going on and we are not aware of that, we are not being explicit about it, then it might very well be that not everyone is saying what needs to be said. And there will be there will be valuable knowledge in what everyone has to say. So we have to be aware of that. And another example here is uh uh is Shadow IT, that's also something that we could do as a reaction on ranking, but can you have a better story about that, right? From personal experience?
[00:23:33] Yeah, exactly, and that and that goes into also the symbolic violence, but there's also an opposite to the symbolic violence and that is that you're blind to your power that you're having. So, if you're in the ranking and you're thinking, okay, we'll need to make decisions, right? I'm going to make a decision. But you're not hearing your environment again, what will happen is that if you set a rule, in this case it was, well, we're only going to use uh Jenkins for uh continuous delivery. Well, that that didn't work with us because on Jenkins or on uh we couldn't use iOS for instance for our app development. So they just say this is what's going to happen and it was already happening, it was a rule, you need to use it, but it wasn't actually working for us. So what did we do? We ran to the shop, bought ourselves a MacBook, put it in the cabinet, hooked it up to the system and now we could do our tests. But that's really dangerous, right, because it can be I I mean I'm not an expert in security, I am security aware but not at all know the dangers of what can happen and it can really, that can really cost you your uh your company maybe even if if someone could hook up to that laptop from the outside and just start spreading to your company. So knowledge is suppressed there. And uh we recently tweeted about it or put a LinkedIn, right? If a manager would be aware of how much knowledge actually was suppressed or not being said or not being put. Well, that will be a top KPI because I can tell you that's probably more than 60, 70% in most meetings that people don't say everything that needs to be said. It doesn't mean everyone needs to say something, so there's a difference. It just meant is all the knowledge there that we need. So what's happening? Uh there's a nice way to spot this. Right? So you there are some patterns, and this comes from Deep Democracy, there's this thing called a sabotage line. And I we presented this last year, I think, at the conference, but the sabotage line are really nice to spot if knowledge is actually being suppressed. And it starts with jokes, at some point sarcastic jokes will pop up and it always goes to the right. Now the thing is, everyone has their favorite one, mine is stop communication or I'm just going to disrupt, that's my favorite sabotage thing. And one one thing that happens on the line isn't per se really bad because well we all have this.
[00:25:55] But as soon as something becomes a pattern, it's nice to spot this. And you see, ah, I think there's something here that's not been said. Maybe we should get this out. And the way you can get it out is ask the question to the group. Hmm, I see that we're slowing down. Do you have the same feeling? And see what happens, let the group solve it again. So, that's what's happening once uh once you create ranking and you're not aware of the ranking and the decision you make or even the the blindness to your ranking that you don't even know that you're suppressing people's knowledge. This will start to happen. So this is the flip side, right? If you're higher on ranking, start having these sabotage line in the back of your mind. And well, since I'm on the other side, right, Evelyn feels mostly on the lower ranking, me on the high ranking, I have this in the back of my mind. Whenever this happens. Ah, sarcastic jokes, maybe I need to pay attention now.
[00:26:52] So one thing also to know is that every autocratic decision, so every decision you make suppresses knowledge and wisdom and creates resistance. on that sabotage or resistance line. Which isn't so bad per se, we're going to get to decision later later on because of course, we need to make a decision, right? But it's just so that you're being aware. Now, if you take this even further, so if you're not aware of your rank, then all of a sudden you might come something into group think, right? We're behaving like each other, we're in a we're in a one way and then at some point you put diversity into the mix. Because you know, diversity works, we copy paste the model. Uh right? So what happens then is this new thing and I would like to see give you 10 seconds and see if you write in chat what is it that you spot here.
[00:27:40] I know the answer.
[00:27:43] Give you 10 seconds, see if someone reacts in the chat.
[00:27:49] Yeah. Yeah.
[00:27:52] Yeah, so this bird is upside down.
[00:27:56] Yeah, everyone.
[00:27:58] Right? And and that's, yeah, that's what we spot. And and that's represents Harry. And it doesn't need to be hurry, but it's there's a whole story behind it. But whatever it is, that's the weird person in the group.
[00:28:11] And we're going to name that person a weird thing. Right? And what's going to happen is we're going to start ignoring Harry because he has he or she has all these weird ideas, right? Uh, but the thing is, what you don't know is, those that were seen dancing were thought to be insane by those who could not hear the music. This is an anonymous quote. Um they don't know where it came from, but this is actually represents Harry in this case, right? Because you don't see what Harry sees. And there's valuable knowledge there, and just saying, ah, this weird person again, now you're suppressing knowledge. And that's really in a collaborative modeling session especially, you don't want to do that. Because what we do is we want to embrace that weirdness, right? If everyone's just trying to be normal, why would we have meetings at all, because then, well, everything would be settled quickly, right? Which is not. And and that's what we want to do. So, what we want to do there and so how to make sure everyone said that how to make sure everyone said what has to be said in this session and battle your ranking. We already said try to create awareness of that rank. So especially if you're in that rank, share that rank and play around with that rank. That's what we mean with playing around, right? Create that bingo, because if I'm the facilitator, I'm higher in rank, so I can start doing that. But also, really important, own your rank. If you are this top rich person or this CTO person and you're saying, well, I'm just I'm just like I'm just like the rest here, right? No, we're equal. No, you're not, you're a C-level person or you're a manager person, you have more power, you have access to budgets and money flow that someone, some developer or some agile coach or whatever it is in the team doesn't have. So never say that you're equal. It's like saying if you're a parent and you say to your small child, no, we're best friends. No, you're not, you're deciding when the child needs to go to bed, you have the power. But, you know, you want to be equal. So it's about equalness, so you want to play and share that uh going into that sharing, that means like if I have a lot of marbles, because, well, I'm I'm a privileged person, then I can share the marbles with someone, right? So that that person can also play in the game. Going back to the marketing uh story of Evelyn, if I would be in that meeting and they will trust me because I'm a tall white guy, then I will deflect all the messages to, hey Evelyn, you know the best of this. And that's how I share my rank. Right? We need to be aware of it, but we also need to own it. It's not bad, it's happening. So see if you know that and Evelyn and me do this all the time, we try to share that work. So that's important. So how to make sure everyone said what has to be said, own, play and share your rank.
[00:30:58] Yeah.
[00:31:00] Right? So that's the the the first question. That we uh that was our first uh question that we solved when it went to ranking. But how can we create a new.
[00:31:09] And that's uh well if you if you've read the title of this uh of this uh presentation and maybe you you already have some idea. Um but we are now going into a new part and I want to start off with a with a very small quiz. Um don't worry, we won't uh we won't rate you or anything. Um but this is to kind of uh introduce the next section. So a man driving with his son beside him gets into a car accident and the father dies instantly, the son is being transferred where ambulance to the hospital. The surgeon walks into the operating room, sees the child and says, I cannot do the operation since this is my son. How is this possible? Um if there are any ideas, please leave them in the chat. Um because this will kind of give us an idea how our Yeah, probably some people already knew it. Um but this is a when you when you don't know this quiz or this question, it is a very hard question to answer because we are biased in a way. We are conditioned, our minds are conditioned, we are living in this world and we uh we have to deal with that. So this is a very much of a cognitive aspect. So I want to go into cognitive bias a little bit.
[00:32:16] Yeah. Yeah.
[00:32:19] And and maybe.
[00:32:20] Give the answer away.
[00:32:21] Yeah, Oh sorry, I talk it publicly available. But yeah, it was a mother.
[00:32:24] Yeah, sorry.
[00:32:25] Yeah.
[00:32:26] No. Yeah, and good thing to notice. This is a study and in the study 80% of the people didn't see it, 70% said they were feminist and a lot of them had mothers that were in a doctor position. So, that goes into.
[00:32:41] Yeah. And that's so much about how we are wired and how our brains work. Um so yeah, let's let's dive into cognitive bias a little bit. Um I'm I'm not doing the whole talk about this because we simply don't have time. Um but cognitive biases are mainly um short mental shortcuts or small mental shortcuts that we use to make sense of the world. So these things are the reason that we don't have to think about every little decision that we have to make each day. So when I would ask you, hey, um what exactly did you every every single choice did you make to uh go from your uh uh from your bed this morning to lunch, could you describe every single decision? Well, that would be a lot. Luckily we have this kind of automatic pilot that helps us there and these are the biases. But even though they are very useful and very helpful, at some points they are very harmful and we should be able to recognize them and uh encounter them. So I want to discuss three biases, three examples that we very often see in collaborative modeling sessions, um and the first one is called functional fixness and I really, really love that that that wording.
[00:33:48] Um but this is basically about getting stuck in what we know. So we have this we have this this this image of the world for example, or uh we sometimes use in our collaborative modeling sessions a fake domain of a uh a cinema, a movie theater. And we have this this mental image of what a cinema should look like. And as soon as we start asking people, hey, could you come up with some solutions for example, for the COVID, the Corona crisis, we have to have one and a half meters in between seats. We still use the image that we have from this cinema, from this movie theater. But maybe if we would think in diagonal rows or maybe we should just remove all the all the seats in there and we should just place couches in there, I mean, you could really go crazy.
[00:34:33] Um you could go against this mental block and because what we know hinders us to take a new perspective. And I really love this image on the right. Um this is actually a recent example I saw on Twitter, so when the rest is small but the owner is creative. They have to, it's small, you have to have distance between the seats, but this is really um functional fixness averted. And this is what I really like. So you when you are in these collaborative modeling sessions, it's really valuable to invite these Harries as can you call them because they will get you the new ideas and the new perspectives and they will counter that functional fiction. So this is something that you very often see in these sessions but it's yeah, it's thinking outside the box and getting those new perspectives on. Uh the second one that I want to discuss is the anchoring effect. And this mainly is about that we think relatively rather than objectively. So the first piece of information that comes to us, that is our source, that is our anchor. And we try to adjust based on that anchor. So, for example, when Kenny and I are preparing this talk, um maybe Kenny said, well, we should be able to prepare this in two days, right? And maybe I would have said, well, I need five days. But because he already said two days, that's the anchor. And I will move towards that anchor. So maybe I will say three or three and a half, I will just move towards that. So when you are in a collaborative modeling session, you can imagine that once someone is saying something, everyone will move towards that anchor. And that could be very harmful because you are missing ideas on the one hand, but you are also going into this this average thing where everyone moves towards each other and you won't have the radical ideas. So this is why we really love to use also individual contributions in the session. So we are using posted so that people can first individually place their ideas on a board. And then we start discussing them so that we are we are not biasing each other and we are not placing anchors where they shouldn't be. And then the last one, the last bias that we see a lot is the false consensus effect. We are almost done with Trump, you're still there, but yeah, I really like uh he's really relatable to this to this bias. But this is basically the overestimating the extent in which others share your beliefs, opinions and enthusiasms. And when you are working on something, when you're collaboratively modeling something, you are putting a lot of effort in there. And um you will start to think that this is the best model the world has ever seen. That doesn't have to be the case, spoiler alert. But once you start working on that and putting effort in it, you will start to think that and you will think that the rest of the world or the rest of your organization will share that belief. So this is why you always need to check your assumptions. You cannot just think that your model is the best. Even though it might be the best, just consider the idea that your ideas might be wrong and that others are not on board with your ideas. So, um in short, uh well, there are there is an extensive list of cognitive biases and I I cannot discuss them all. Um but I will I will put a link on on Twitter after this talk. But you have to be able to at least recognize some of them to counter them. So, um how can we create and include new insights? Well, we need to embrace the Harry's uh to counter the functional fixness, we need to drop our anchors consciously. And we need to leave room for individual contribution, um so that we won't move towards each other and yeah, get stuck with a very average solution. And we need to identify and check our assumptions and make them explicit to counter that false consensus effect, and that means that we really, really have to consider the possibility that we might be wrong. And that is a very hard thing to do, I know, um but it helps. So collaborative modeling is very good but leave room for individual contributions preferably at the start and then you can merge the ideas. So yeah, this is basically um how we uh how we deal with that.
[00:38:18] Yeah.
[00:38:19] Yeah, very.
[00:38:20] So now it's easier, right? We just decide what we have ton more information, now it gets easier to decide. Right? So that's how you you gain. So if you're being aware of your ranking and you're being aware of your cognitive bias especially during this collaborative modeling. Now we have everything there to make a decision, right? Actually it gets harder.
[00:38:41] Right? Now, now we're at the point that we have so many information, what are we actually going to do? Um So, uh now you're getting going to get next problem and that's now you have maybe too many diversity or too many options. So which one do you pick? Question to the audience, which one do you pick? Do we actually have all the information there? That's the question as well, but which one does the audience pick? Under or over?
[00:39:05] Hey, over.
[00:39:07] Yeah.
[00:39:07] Hey, over.
[00:39:08] Yeah.
[00:39:09] Oh, well, over.
[00:39:11] That's the over one, right? So actually, yeah, there's a pattern here that says you need to go over. But actually I have a cat at home that if it attacks the the roll, under is even better, right? So still now, we're focusing still maybe on solutions. And solutions are not the why. And the reason I'm saying that is if we really want to understand each other, because this is what's going to happen in if you have too many options is that it's going to you're going to get groups, right? And you're going to have so many options. And now the groups are going to get fixed on these solutions. Because this is a solution. But why do you actually want to have over or under? That's the important question. And to do that, we actually need to lose the other, we need to lose our solution. We need to make sure our solution is gone and can listen to all the other inputs truly. So, this is the difference between accurate versus completeness. Usually we try for accurate, but we want to have completeness. So in if you see the duck, in order to see the rabbit, you need to see the you need to unsee the duck. Right? It's not possible for me to see the rabbit from seeing the duck. And this one is easy, but let's go to colors.
[00:40:24] Right? I see gold and blue, some I think see black, right? I cannot see black in this picture, it's impossible for me, really at this moment to do it. And that's also in the social part, right? And um what we're doing in these dialogues, in these collaborative modeling is that especially in the Western world, we're so fixed in discussing and not in having a dialogue.
[00:40:00] I can listen to all the other inputs truly. So, this is the difference between accurate versus completeness. Usually we try for accurate, but we want to have completeness. So, if you see the duck, in order to see the rabbit, you need to see the you need to unsee the duck. Right? It's not possible for me to see the rabbit if I'm seeing the duck. And this one is easy, but let's go to colors. Right? I see gold and blue. Some I think see black, right? I cannot see black in this picture. It's impossible for me, really, at this moment to do it. And that's also in the social part, right? And um what we're doing in these dialogues, in this collaborative modeling is that especially in the Western world, we're so fixed in discussing and not in having a dialogue. We're not, we're not so we're too many fixated on winning the discussion. Instead, what we need to do is having a dialogue first. There's and we got to get to that there's no there's no better one. But in the Western world, we're so locked up into the discussion, which is about actions, thinking fast, convincing, take a stand. you know, see the ranking playing again, opinions, arguments, attack, defense. While to be complete, we need to have insights, we need to understand and listen, think together with others. Yes, and the questions are central. So why are you wanting this and not the solution? So not the opinions and arguments and the solutions. Because what you're getting is these multiple perceptions now, because you're invited in, which makes sense because we can learn a lot from them. But we also now need to switch how we're dealing with that.
[00:41:38] And um what will also happen is there starts to be conflict, because it's really hard to see each other's standpoint. It's actually quite hard to really understand the other person, to see the other perception, it's really hard. And people often in the Western world also see uh we want to have peace, we're really conflict avoidance. especially in the Netherlands, I can say, we're conflict avoidance. But actually, if you look at it, is that peace is not the absence of conflict, conflict is the status quo. Conflict is always everywhere. Just as a system is always near failure, uh there's always conflict. But it's how you manage them is what makes peace. Right? And that's what we're forgetting in companies. If we're bringing in this diversity, we're copy-pasting again. Conflict is good because if we solve the conflict, we learn. Just as from failure you can learn, from conflict you can learn as well. Bringing in all this diversity, you need to have managers and people in position that knows how to deal with conflicts. Else, it will be a self-fulfilling prophecy, and diversity will never work because we're not dealing with the conflict. And a little bit here is about truth. And here's the thing, the why usually is we're pretty similar on what we want. And conflict always comes because we want the same. But the why here is that we also are really rush to other diversities. But what we actually want are five basic things: clarity, feelings of others, own image, space, effectiveness. It's just how we implement them are different. For instance, clarity maybe in my world, clarity means I'm going to talk for 40 minutes in this uh in this session, while my wife will say, no, I can go with one one sentence and you know what I mean. Right? So we want to have the same, it's just how we implement it this is different. And there's where the conflict starts to arise. And that's where we need to deal with. So in order in these collaborative sessions, we talked about this before, is in order to get everyone's perception really in, we need to understand each other, but that means we really need to truly listen to what's being said. And all the parts in the group, even those the ones they don't like or believe to be useless, must be present and supported. And supported means the person needs to be felt, listened to. Right? So that's really important. The way we start this is we need to stop having these meetings. We need to have more campfires. And with campfires, I mean, we need to see each other in person. We need to discuss personal things first. Because there is where the connection is. And there's a really nice movie, it's called unboxing, where they bring in multiple groups of stereotype and then they're going to ask these why questions. So who has done this and then the group started to unbox and started to go together. And that's in these meeting is really powerful to do, and from Louis Deep Democracy, that's what you're doing because you have to group the conscious the consciousness and the unconscious. And the wisdom is in the unconsciousness. You want to bring that up, that means we need to spread around the things that people have in common. And the way you do that, uh there's a nice tech my steps in deep democracy and uh the first step is gain all views. Meaning we're going to ask people, okay, what is your view? What's your perception? And people really need to listen to each other. And not listen to react, but really listen. Make it safe to say alternative views. So I always ask the question in this heuristic, okay, but who sees it different? And give that person space to talk. Then we're going to spread the why. And here I got a really nice example of a uh colleague of mine that says, well, my child came uh wanted to go to school in the morning in her pajamas. Well, you cannot go to school in your pajamas, right? So, uh someone might say, no, go get on a pants. But what she did was ask the question, why? And the why was, well, I want to have a comfortable trousers. Aha. So that's your need. So let's focus on the need and spread the need. Right? If you just focusing on the solution like over or under, that's not possible. Let's focus on the why. Why I need under is because I have a cat that screws up the system. And let's see if other people have the same feeling, right? I'm not focusing on the who wants under, who wants because that's about discussing, I'm focusing on the why. Then we're voting in deep democracy and here's where the decision making comes in, right? We're voting and since it's a democracy in deep democracy, that means you need to have a majority vote. But number four is the most important thing. Ask, what would it take for you to go along? So people who hasn't voted for the majority vote, what would it take for you to go along? A really powerful questions. And just just don't follow these steps, but just do four, right? Once you have a once you have made a decision, always ask the question, what would it take for you to go along? That already in itself is really powerful. Now, that's how the deep democracy, the Louis method steps, so deciding an architecture can be done in this deep democracy. And there's more to it. You can learn more about this. But one thing I want to give one thing I want to give away and I already said give space. But also be sure how much space you give. So there's four ways that you can go into a decision-making. The first thing is, just I have an idea, let's try something. Uh and everyone, everything is open for discussion. The second one, well, I I I got something said. I have a suggestion, maybe we want to go on a trip to, well, I don't know, I want to uh I want to use this new programming language. Right? It's already a suggestion, you have a clear intentions, but other insights are more than welcome. And then you get to the proposal. You have a concrete worked out proposal, only serious objectives can influence the decision. And four, still needed, we still have big companies that we have hierarchical decision, which is the command, but what does it take for you to go along? A question that's never been said by a sea level or a manager. Because the door is always open, right? Nobody will go to the door. So, um,
[00:47:48] one thing here, how much space do you do you give? One thing that's important here, be clear how much space you give. Because sometimes people just get in and they're like, oh, everything is open, right? We can do collaborative modeling and I can decide. While actually, I'm coming in with a proposal. So you're giving people a false consensus about how much space they have. So these two things are really important when you go into decision making, right? So who decides on the architecture? Be clear who decides and how much space and how much room you have. And how do you get everyone on board on the decision is ask that question. And really listen to people. People don't really uh find it uh hard to accept a decision as long as they're being heard with their problems, right? So we use the deep democracy to the Luis method steps and be clear in how much power we people decision making.
[00:48:41] So now, the last question, who to invite and how diverse should the group be? So now we made a decision but a thing that people still is, okay, who do we invite and how diverse should it be? So, there's not a really good answer here, but there's a few things we like you to be aware of. And the first thing is, exclusion comes from having too much inclusion, and inclusion always leads to exclusion. Be aware of it.
[00:49:08] You always exclude. Just having that in your back of your mind, I always excluding, okay, who are you excluding and how are you dealing with that exclusion? How are you managing that you're excluding people and does that make sense? And have that conversation with that person. And and there's the paradox here in inclusion, right? Because we want to have harmony. But you got to be yourself, but you also need to act accordingly. And that's torn, right? You're torn between these two worlds. Okay, I need to be myself, I need to act, but I also need to to be myself. And that's that's the split between um the paradox that are here, and we call that polarities. And that's the thing when we think about polarities, for instance, happy or sad, well, I can be happy and sad. And that's the thing we want to give you uh we give you away is that polarity thinking. We need to have and.
[00:49:59] Yeah, absolutely. This is uh this is a very important one. Uh we have this this tendency uh that we we want to solve problems. And of course, we want to solve problems. That's what we do, that's who we are. But not everything, not every problem is a problem that needs to be solved. Some conflicts are not really problems that we need to solve, but they are polarities that we need to manage. Um and that is a very important distinction. So we need to go from that either or thinking to both and thinking. And that's also very important when we invite certain people to this collaborative modeling session. So as this nice quote says, it's who to invite is about thinking and perceptions instead of people. Invite especially people with annoying perceptions. So the Hari, uh we should definitely invite them. And this will trigger these polarities uh that we can manage as a group. And that is what we uh uh what we need to do because there will always be polarities in what we are doing. And um well, for example, well, this is my all-time favorite book, if you have some time, please read this. Um but for example, there will always be people there will always be people in a room that want to go really fast and I want to be people, there will be people who want to go more slowly in uh in this collaborative modeling session. Well, it's not either or, right? Um this should be both and thinking. And this is not a problem that has a clear answer. It's not a clear problem such as what do we have for dinner? That's a very easy thing to solve. Well, most cases it's a very easy thing to solve. But um this fast and slow, for example, teams versus individuals, those are polarities that we need to manage in a way. And we have some uh some tools for that, which we also uh we are doing some workshops on this and it's it's a really, really cool tool that you can use. And it's called a polarity map. Um and we will uh uhm yeah, given the time, I will explain this one.
[00:51:49] Yeah.
[00:51:50] No, that's uh that's sorry.
[00:51:53] Yeah, sorry, I see the other slides now.
[00:51:54] That's why. So what we what we have here is we have these two polarities at the left and the uh left and the right side. So team and individuals, for example. And this this thing will force us to come up with positive and negative effects of both of these sides. And when you do this within a group, you will be forced to see multiple perspectives. You will be forced to go through this polarity map in this infinite infinity circle uh uh thing. I don't have the the cursor so can he maybe you could show the how we walk through this map. Um but this is this is a very helpful tool that we that we are using at this point. Um and it helps you to see these different perspectives, it helps you to see the other side and it will also help you to convince others or to get the majority on board with your decision making.
[00:52:46] Yeah, and a good thing there is, um, so you do pick both sides here, but what's really interesting is that you look at the actions that you can take to go to the other side, and what are the signals or observation to see when do we go to the negative.
[00:52:59] Yeah, I'm not seeing the slide anymore, so you can uh.
[00:53:03] You're trying to stay in the positive, right? Sorry.
[00:53:09] Oh yeah, sorry. So what you're trying to do is always stay in the positive. It's like breathing in, breathing out, right? I always want to stay in the positive and I don't want to lose too much oxygen or get whatever it's called. So I'm trying to stay here. So if you stay here and you're trying to be a cohesive unit, common direction, team support, all of a sudden you will start neglecting personal needs. So once you see that, okay, then we need to go to the individual, address that personal needs, right? At some point, when you're staying in this side, what will happen is that you start becoming isolated. Okay, and then we need to go back to the positive effects. And just the most of the discussion that goes into these polarities, uh isn't one to solve, right? Oh, then we do again a team meeting. Oh, then we do, let's go back to solo meeting. Just having the polarity mapping makes it complete and then we can decide on actions and signals once we move through the both of polarity. So in collaborative modeling, you always see these polarities uh starting to behave.
[00:54:10] So I hope that was helpful. So for us that that means we're going to do these polarity mappings during the sessions if we see a big polarity popping up, let's map it and let's decide together, okay? One thing in collaborative modeling what you see is, okay, we don't have new insights. Okay, let's switch to isolation. Great, and start managing that actively.
[00:54:29] Oh, now I'm only seeing myself. I'm not even seeing you anymore.
[00:54:31] So you still have no slides, have you?
[00:54:33] Sorry.
[00:54:34] Yeah. Good. So, uh we're at the end anyway. So, uh ensuring flow in meetings. So create awareness of people that ranks in the group. Own, play and share your rank. Be aware and make explicit the biases at play, play with that biases, right? Use deep democracy in your decision making, especially keep asking, okay, who feels the same way on this why, um, and what do you need to go along, and start managing polarities with the group by using polarity mapping. And don't let it circle around in these meetings, just dive into that polarity together.
[00:55:09] So now I'm missing the cat.
[00:55:10] And uh
[00:55:11] I'm missing the picture of the cat.
[00:55:12] There was our talk.
[00:55:13] Yeah.
[00:55:14] Have some contacts.
[00:55:15] Now you're missing your cat.

Transcript