Skip to content

Coherence Busting explained

In Action Science workshops I teach a technique I call Coherence Busting. To understand why it is useful I ask the audience to imagine themselves making a proposal: “While you are talking you notice the main stakeholder — the person in the audience you most hope to persuade — glance at their watch. What do you do?”

I ask this question to allow the audience to experience the decision-making heuristics that Daniel Kahneman describes in this book Thinking, Fast & Slow. Kahneman models our consciousness as being two systems, our fast, automatic, unconscious System 1 and our slow, deliberate, effortful System 2. Part of what makes System 1 fast are the shortcuts it uses. Two of these shortcuts consistently arise with the watch example. The first is that we  assume that a coherent story must be correct. The second is that we limit the facts to what we can immediately recall, a process Kahneman calls What You See Is All There Is (WYSIATI).

These two shortcuts are displayed in the watch scenario.

We unconsciously construct a coherent story for what the glance means — for instance “she has somewhere else to go”. This story is based on first thoughts of what the glance might mean (WYSIATI). The coherence in our story give us the sense that our story is true. We then design our actions in response to a story we made up. This is the key lesson of the watch example: We feel as though we are responding to the reality of the situation, because WYSIATI and coherence cause us to mistake our single plausible story for the truth.

This is why we need Coherence Busting.

With the watch example, I ask the audience to describe what they think the glance means. After I have harvested the normal stories from the audience (“they’re bored”, “their attention has drifted”, “they are running short of time”) I ask them to consider other possible meanings of the glance, all of the possible reasons, even wildly implausible ones (“they have their plan for world domination written on their hand”). Now the audience generates dozens of possible reasons: it is a nervous habit, they were admiring their new watch, there could have been an alert on a smartwatch, maybe an itch on their wrist, and lots more. What makes this Coherence Busting is not just that there are lots of options but that the options are mutually incompatible. Once we can imagine conflicting explanations we are not longer trapped by the original coherent story. These options were always there, but it requires invoking System 2, our conscious and effortful thought process, to bring them to the surface. That’s not something we do when we feel we already have a good explanation. So what would trigger us to use Coherence Busting?

I’ve found Coherence Busting a useful tool to reach for when I recognize that I’m frustrated. A common pattern for me is to get frustrated when I can’t come up with a justifiable explanation for the other person’s actions, when I don’t like the explanation that System 1 has suggested for me. When I recognize that pattern I try and think of at least three incompatible motivations for why they might be behaving the way they are. The technique of Coherence Busting is a way of reminding myself that there are infinitely more possibilities than I have considered. It allows me to let go of the story I’ve made up about the other person. It reminds me that if I want to understand what the other person is thinking, I’m going to have to get out of my head and into theirs — probably starting with asking them a genuine question about what they are thinking.

Much of my work with Action Science is learning the skill of asking good questions. Coherence Busting reminds me to use those skills.

(Thanks to Douglas Squirrel for helping develop Coherence Busting. See more of our work together at and the London Action Science Meetup.)

Got “museum sleepies”? Illustrating mastery, learning and ego depletion

At a certain point in reading Daniel Kahneman’s Thinking, Fast & Slow I realized I had discovered a possible explanation for the mystery of “museum sleepies”. Museum sleepies is my wife‘s term for the fatigue we feel after a rather short time in a museum, a term we’ve used much more frequently since moving to London. I know this is a common experience, and a search will find various explanations, both physical and mental. What I got from Kahneman is a better model of what’s happening to us, and — very exciting to me as a learning nerd — a testable prediction.

The focus of Kahneman’s work is decision theory, and the title mirrors the model of cognition he uses. We can imagine that our mind has two systems for processing the world and making decisions. System 1 is fast and intuitive (think Blink), unconscious, involuntary, and cognitively cheap. System 2 is our deliberate, conscious, analytical thought process that is also, unfortunately, both slow and cognitively expensive. Most of the book is an explanation of how our reliance on System 1 results in predictable cognitive biases. To get there Kahneman first describes experiments to establish a measure of mental effort, evidence that all types of mental effort draw on the same limited pool of resources, and that engaging our System 2, focusing our attention and performing deliberate analysis, draws off that limited pool.

In the larger world the findings Kahneman describes have serious implications in terms of ego depletion and decision fatigue. In our tour of the museum there isn’t much at risk, but it struck me as a application of the same findings. After reading Thinking, Fast & Slow my model for the museum sleepies is that I’m using my System 2 to analyze one artifact after another and that this is draining my limited pool of mental energy. This model leads me to predict that someone trained to analyze the artifacts wouldn’t suffer the same effects. The trained person would see the same artifacts differently because they would have a set of prebuilt patterns and categories to draw upon. They would notice the patterns largely through System 1; System 2 efforts would be brief and efficient. Finally, I expect I could use deliberate practice to train myself to understand some class of artifacts, and that henceforth I would no longer suffer the same fatigue when viewing that type of collection.

I’ve spent a lot of effort in the past few years trying to develop my abilities to promote organizational learning. Much of my focus has been on the Action Science approach to effective communication. Another part of it has been trying to teach the scientific mindset that learning is about detecting and correcting error, which we can do by making testable hypothesis and testing them. I don’t plan to conduct any cognitive psychology experiments with art historians, curators, and artists, but I love the fact that someone could. This museum sleepies story has become part of my arsenal for illustrating some important concepts in cognition, in training, and in theory making. I hope you find it useful too.

Pretend they are mind reading

At the prompting of Douglas Squirrel I just read Yossi Kreinin‘s blog post People can read their manager’s mind. This seemingly magical power is the mundane result of combining “People generally don’t do what they’re told, but what they expect to be rewarded for”, and people are good at spotting what is actually rewarded. As a manager/leader I’m taking away a heuristic I want to test:  when I’m not able to get alignment with my stated goals I’m going to pretend the team is reading my mind, and that they are heeding my hidden thoughts rather than my words. From that mindset, what does that suggest about my own values? The results aren’t likely to be flattering. I embrace the idea that “the only way to deal with the problems I cause is an honest journey into the depths of my own rotten mind.

(As much as I embrace that message for myself, I’d warn non-managers from seeking solace in the article, from using it as a shield to deny their own accountability. Yes, the article says that it is an “insane employee” who will work to fix important but unglamorous problems, and that “the average person cannot comprehend the motivation of someone attempting such a feat”. Do you find solace in being average? In being powerless? I don’t. I think it is worth always seeking to improve, and to improve the organization I’m part of. I believe someone out there could improve the situation I’m in, so if I’m frustrated it is probably my fault.)

Playing to Win in Software

Subtitle: Lessons learned from video games

This past year I’ve spent a lot of my time on Action Science as a route to organizational learning, and one of the real insights I’ve had is how painful it is to learn. Learning, according to Chris Argyris, is the detection and correction of error. The emotionally difficult part is detecting our own errors, in being genuinely open to the idea that we are part of why things aren’t working better than they are. In my view a professional should cultivate the mindset that they need to improve. Otherwise they risk being a “scrub”.

Slides (pdf): Playing to Win in Software

This talk was delivered January 10th, 2013 at TIM Group as part of their Thursday lightning talk series.



That’s what learning feels like

None of us like to be wrong. I’ve tested this with many audiences, asking them “how does it feel when you’re wrong?” “Embarrassing”, “humiliating” or simply “bad” are among the most common answers. Stop now and try and think of your own list of words to describe the feeling of being wrong.

These common and universally negative answers are great from a teaching perspective, because they are answers to the wrong question. “Bad” isn’t how you feel when you’re wrong; it’s how it feels when you discover you were wrong! Being wrong feels exactly like being right. This question and this insight come from Kathryn Schulz’s TED Talk, On being wrong. Schulz talks about the “internal sense of rightness” we feel, and the problems that result. I think there’s a puzzle here: we’ve all had the experience of being certain while also being wrong. If the results are “embarrassing”, why do we continue to trust our internal feeling of certainty?

My answer comes from Thinking Fast & Slow. That sense of certainty comes from our System 1, the fast, intuitive, pattern recognition part of our brain. We operate most of our lives listening to System 1. It is what allows us to brush our teeth, cross a street, navigate our way through a dinner party. It is the first filter for everything we see and hear. It is how we make sense of the world. We trust our sense of certainty because System 1 is the origin of most of our impulses and actions. If we couldn’t trust System 1, if we had to double check everything with the slow expensive analytical System 2, we would be paralyzed. So we need our System 1 and we need the sense of certainty it provides. We also need to be aware it can lead us astray.

When our sense of being right guides us we are acting from a Model I / Unilateral Control mindset. The result of a Unilateral Control mindset is less information, reduced trust and fewer opportunities to learn. And we all like to learn right? I now ask my audiences this question and I get universal nods. We all like to learn. “No you don’t”, I reply. “You just told me that the feeling of becoming aware you were wrong feels bad! Well guess what? That’s what learning feels like.” That’s my recent ah-ha moment: that we claim we like to learn, but when it actually comes to learning, to correcting a wrong belief with a right one, we don’t like it.

I find this discrepancy very interesting, very revealing. My theory is that when we imagine learning we are thinking of writing on a blank slate. It is about learning facts where before there were none. That is a good feeling, we get a little chemical kick from our brain when that happens. We don’t imagine correcting our mistaken beliefs when we think of learning, and that’s a real shame, that should change. By all rights we should value that kind of learning even more than learning new facts: “It isn’t what we don’t know that gives us trouble, it’s what we know that ain’t so.” (Will Rogers)

I think the problem is that we are primates. To primates, from an evolutionary psychology standpoint, status is everything. Status is the primary determinant of reproductive success. Losing status can be the same as a reproductive, evolutionary death sentence. In our modern knowledge economy, chest thumping is the assertion that we are right, and winning the fight is proving the other person wrong. That’s how we put them in their place (in the status hierarchy). This means our instinctive reaction to becoming less wrong tends to be negative. The loss of status feels too high a price to pay for learning. Even trying to help someone else become less wrong is understood as a risky prospect. We don’t want them to lose face, we don’t want them to get angry with us for correcting them. Thus the habits of Unilateral Control, protecting both ourselves and others, are reinforced.

All of this explains why developing habits for learning, developing Model II / Mutual Learning habits, requires a lot of practice. We are fighting decades of acculturation on top of millions of years of evolution. To win this fight we need to be committed to what we are fighting for. We need to care more about learning than being right. We’ve got to care about making the most informed choice possible. When I can remember to hold these values in mind it becomes easier to act differently. I can go seek out those people who are most likely to disagree with me, who are most likely to teach me something. I can deliberately share my chain of reasoning and invite others to poke holes in it. With practice, lots of practice, I can come to see the person who corrects me as more friend than rival, and to feel the correction as the victory of joint learning rather than an individual moment of shame.

“The story I’m making up…”

Last week at the August session of the London Action Science Meetup we started with a discussion of the phrase “the story I’m making up…” I love this phrase! It captures the process of the Ladder of Inference, but it has an immediate emotional resonance that the ladder does not.

I came across this phrase from an unlikely source: Now I’ve got nothing against Oprah, but this it just isn’t in a url that shows up a lot in my browser history! So the real credit goes to @frauparker, who is someone I clearly trust well enough to follow off into the well-scrubbed bright and shiny parts of the internet; in this case an article by Brené Brown on How to Reckon with Emotion and Change Your Narrative, an excerpt from her new book Rising Strong. The article is well worth reading but I’m going to take a look at it through the narrow lens of the action science workshops I’ve been leading.

For a few months now I’ve been leading weekly lunchtime workshops at TIM Group where anyone who is interested can join in a discussion of some two-column case study. This might be a canned case study, or one people recall and write down on the spot, but the best sessions have been when someone has come with something in mind, some conversation in the past or future that is bothering them. As a group then we try to help that person to explore how they were feeling, what they didn’t share and why, and how they might be more effective in the future. One of the common observations, and one this article reminded me of, is how hesitant the participants are to put their feelings into words. It seems we can almost always have a productive conversation by asking the series “How did you feel at that point?” followed by “Is there any reason you didn’t share that reaction?” These questions get quickly to the real source of our frustrations that are hiding back in the shadows. Bringing them into the light we find out something quite surprising: they were stories we were making up.

The article has good advice that will be familiar to students of action science and the mutual learning model, suggestions like separating facts from assumptions, and asking “what part did I play?” It also had a key action that is hidden in plain sight. Do you see it in this exchange from the article?

Steve opened the refrigerator and sighed. “We have no groceries. Not even lunch meat.” I shot back, “I’m doing the best I can. You can shop, too!” “I know,” he said in a measured voice. “I do it every week. What’s going on?”

As I read the remainder of the article from the author’s point of view it was easy to overlook the role Steve played here. It was his calm empathetic question that allowed the author to reply with the title phrase: “The story I’m making up is that you were blaming me for not having groceries, that I was screwing up.” I really struggle generating this kind of response, of not getting caught up with my own emotional reaction. So I admire what Steve accomplished here.

I’ve long said that “stories are the unit of idea transmission”. What I hadn’t realised before reading this article was how powerful I’d find using the word story to describe what is going on in my head. I’m really looking forward to practicing with this phrase in future conversations.

Video: Frustrated? It is probably your fault

I had great fun with my Devopsdays Amsterdam talk and the video is now online, joining the slides that I posted previously.

The last time I spoke at a Devopsdays was London 2013 (video here). That was another fun talk, and had some overlap in content, but I did feel that I tried to put too many concepts in a single 30 minute talk. My goal this time was to be much more deliberate and leave enough time for each concept. Where I ended up is a talk in three parts. Part one is cognitive psychology, how our mind generates an illusion of certainty where we don’t deserve it. Part two is Action Science and the Mutual Learning Model as a set of behaviours that appropriate for an uncertain world. Part three is the need for practice. This section uses the piano analogy and then — my big risk! — a live demonstration. A brave member of the audience joined me on stage to try applying the concepts I’d just discussed.

As you sit and watch this video I hope it is the final of section of the talk that makes the biggest impression. All the video watching, all the reading, all the learning will mean nothing if you don’t act, if you don’t practice and find the limits of your current abilities and then learn to move beyond them. And I’ll make the first step easy: Download the slides and try the exercise in the video. What would you say to Ted? Write it out and read it aloud. How did you do? Maybe you want to try again…

(Lots more great talks online at the devopsdays vimeo account.)

Slides: Frustrated? It is probably your fault

I’ve just delivered my talk Frustrated? It is probably your fault at Devopsdays Amsterdam. That means my slides must be finished! Here they are to download, and I’m looking forward to the video being posted later.

The piano analogy: some practice required

This year I’m training people in the theories of Chris Argyris, helping them to apply the concepts, and this raised some fun challenges. The challenge on my mind today is how to convince people that practice will be required before they can perform well? My current analogy is the piano.

After a quick search I can show you a three minute video of a 14-year old explaining how a grand piano works. If you’ve got an extra minute I could share a four minute animation that illustrates the mechanism in detail. You probably already know that in a piano the strings vibrate and that produces the sounds you hear. It would take moments to strike each key and allow you to hear each note. Having invested less than thirty minutes you could understand a piano and how it works. You can’t play it, but you know you can’t play it. You were unlikely to have mistaken understanding the concepts for being able to produce the result.

Action Science seems different.

I’ve introduced dozens of people to the topic through Roger Schwartz’s excellent Eight Behaviours for Smarter Teams, a sort of Shu-level guide to producing Mutual Learning behaviour. The response is typically positive, enthusiastic even, and general  agreement they should start behaving in a mutual learning way. However they also believe that now they understand mutual learning behaviour they can also produce mutual learning behaviour. They mistake understanding the concepts with being able to produce the result. Worse, their own incompetence makes them blind to their lack of skill.

So this is where the piano analogy comes into play. Everyone acknowledges the gap between understanding and performance. I use the piano analogy to set the expectation that practice will be required. Then we begin using the two-column case study to start retraining their ear, allowing them to begin hearing the difference between Model 1 and Model 2 behaviour for the first time. And when someone is discouraged by their performance, the analogy is there again to help them have realistic expectations: “How long have you been practicing the mutual learning approach? How long do you think it should take to retrain from a lifetime of habit and cultural norms?”

Do you have a technique you use to help set expectations for skill acquisition and maintaining motivation? If so I’d love to hear about it in the comments or on Twitter.

Frustrated? It is probably your fault.

That’s the title of my talk which has been accepted to the program for the upcoming Devopsdays Amsterdam June 24th, 25th and 26th. The topics I’ll be discussing will be familiar to the attendees of the London Action Science meetup — and to the people who have been diligently reading my session notes — which is that if you want to create change you need to start by changing your behaviour.

For the past few CITCONs I have been leading open space sessions with a very similar theme: “Can’t create change? It is probably your fault”. Those sessions have been great fun, because we get to talk through situations in real time together. My challenge now is how to generate those same kind of ah-ha moments for the audience without building it together in the room.

I’m not sure how I’ll do that yet, but I’m excited to have been accepted and to have the opportunity to give it a go. Hope to see you there!