Last time, I promised not only to explain how to create the perfect comedy sketch, but also to explain how that knowledge could be used to improve the betterment of mankind, even to the extent of helping us design traffic systems that minimize accidents or avoiding nuclear war. To learn more, read on.
First, let's focus on sketches.
We left off having explored why the brain likes the number three, and why you can't do the same funny thing three times in a scene without changing it and have it still work. What we didn't cover was why the best sketches come in four parts, instead of three.
The answer is, because great sketches have great endings.
You may recall that last time, I proposed that great sketches have the following structure.
1: Surprise
2: Coincidence
3: Pattern.
4: Subversion of pattern.
What people often miss is that the big laughs in a really good sketch don't come right at the end, like the punchline of a joke. They are distributed throughout. We establish a funny, repeating routine early on and then we escalate it, raising the tension the entire time.
However, if a sketch is going to be memorable, it needs to end well rather than just allowing that sense of drama to deflate. This means that at the end of the sketch, you have to find a way to subvert the routine such that it can't persist any further. Ideally, you release the tension in the scene in such a way that it surprises the audience just as much as the material that's come before. In practice, this can be hard to do, but there's usually a way. The easiest approaches are often to look for a way to invert the risks established in the scene, or to reveal some new information that changes the relationship.
Here, then, is a list of sketch components I dwell on when I'm coaching scene-work.
1: Create a character and an environment. Establish the character's opinion about that environment. Try to reveal motivation and vulnerability immediately.
2: Introduce a second character who functions as a foil to the first. Establish a dynamic between the two of them that establishes an obstructed desire for one character as efficiently as possible. The character with the desire is your vehicle for pathos. Your character who obstructs is the vehicle for absurdity.
3: Demonstrate the obstructed desire clearly in a simple pattern of interaction. This can often be done with just two lines of dialog, as counted from the opening of the scene. This can be as simple as:
A: The parrot you sold me is dead, I want to return it.
B: It's not dead, it's resting.
Voila, we have a setting, a motivation, an obstructed desire, and the seeds of lunacy.
4: Repeat the pattern with increased tension. Try to retain as many features of the original interaction as possible, while letting the characters appear to be innovating to try to resolve the conflict.
5: Escalate the tension to the point at which it is transparently absurd by having an interaction that is outside the bounds of normal behavior for your characters. Meanwhile, keep trying to minimize the amount of new content that you add to the scene.
6: Having made the absurdity unmissable, invert the tension in the scene by providing a way out of the situation that the audience isn't expecting. This can be done by revealing an alternative that was hidden from the audience, or by having characters pick a solution that the audience would never choose.
That's it. Spelled out like this, it might seem a little obvious, but it's astonishing how few improv scenes manage to adhere to this level of structure. That's because holding all this in your head while making it up in front of a live audience requires practice. However, it's a tremendously satisfying pattern to have mastered. It provides you with enough solid, funny scenes that you can afford to experiment with quirkier stuff while still having your show still wow audiences.
This pattern for sketches is so robust, in fact, that it appears in a huge number of different contexts, other than those that are simply funny. For instance, it provides the critical underpinnings of the Hero's Journey, the storytelling pattern that gave rise to movies like The Matrix and Star Wars. And, as I alluded to at the opening of this post, it can make the difference diplomacy and violence in confrontation situations, and determine whether a child decides to bother doing their math homework. In the next post of this sequence, I'll explain why.
Wednesday, May 30, 2012
Thursday, May 24, 2012
Comedy and Cognition
Over the last few weeks, I've had the good fortune to be working with the Senseless Bureau, an excellent Oakland-based improv troupe. I've been coaching them on scene-work. Unusually for me, the focus of that coaching has been short-form improv--individual sketches. That process has encouraged me to condense my thinking about what makes great comedy, what it has to do with human cognition, and how we can apply that knowledge elsewhere.
My main conclusion is this: that almost all really strong sketches have the same format, regardless of the humor employed. Furthermore, this structure is a direct consequence of the way that the human mind is wired for learning. I'm also going to propose that understanding the structure of a good sketch can help us build better educational tools, improve computer interface design, and even design road systems that reduce accident fatalities.
What is this magic structure, you may ask, that has so many beneficial effects? It goes like this:
1: Surprise.
2: Coincidence.
3: Pattern.
4: Subversion of pattern.
This probably isn't terribly meaningful in isolation, so let me explain.
The brain, as I've mentioned before, is a prediction machine. We're designed to seek out reliably occurring patterns in the world, and to use them to build the mental models that define our reality. This happens at every level of our cognitive activity, from watching how objects move when we touch them, to anticipating chess moves.
The minimum number of learning instances with a similar outcome that the brain needs to identify a new pattern is three. One new experience is a surprise, but it's hard to know whether any similar experience will happen again. Two experiences is better, but it's still unclear what those experiences have in common. Three experiences allows the brain to rule out noise, and make reliable predictions about future events.
This is not to say that people can't learn from a single experience. We do that all the time. But as you have probably noticed, drawing conclusions from just one or two events is fraught with error. Very often the wrong lesson is learned.
This minimum requirement for new rules manifests in a variety of ways. For instance, in children's stories. Having three instances of something, be it goats, bears, or trips up a beanstalk, is a ubiquitous device, because it feels natural. The same is true in comedy.
As the excellent, and extremely funny Kate Offer once pointed out to me, you can't do the same thing in musical comedy three times. Do something funny once in a song, and the audience will laugh. Do it twice and the audience will love you for it--you've cued up their brains to think they know what's going on. Do it the same way a third time and the funny disappears. This is because by the time you've got to the end of learning experience 2, people are already projecting. To make the third iteration funny, you have to put a twist on your original gag which breaks expectations. If you do that, the audience will love you even more. You've shown them a meaningful pattern, but not the one they were expecting.
So if three iterations is so important, why am I proposing four steps for the perfect sketch? Doesn't 'subversion of pattern' count as step three? And how do we actually use these steps to craft good comedy? Any guesses?
I've probably said enough for one blog post, so I'll have to tell you next time. Meanwhile, all conjectures are welcome.
My main conclusion is this: that almost all really strong sketches have the same format, regardless of the humor employed. Furthermore, this structure is a direct consequence of the way that the human mind is wired for learning. I'm also going to propose that understanding the structure of a good sketch can help us build better educational tools, improve computer interface design, and even design road systems that reduce accident fatalities.
What is this magic structure, you may ask, that has so many beneficial effects? It goes like this:
1: Surprise.
2: Coincidence.
3: Pattern.
4: Subversion of pattern.
This probably isn't terribly meaningful in isolation, so let me explain.
The brain, as I've mentioned before, is a prediction machine. We're designed to seek out reliably occurring patterns in the world, and to use them to build the mental models that define our reality. This happens at every level of our cognitive activity, from watching how objects move when we touch them, to anticipating chess moves.
The minimum number of learning instances with a similar outcome that the brain needs to identify a new pattern is three. One new experience is a surprise, but it's hard to know whether any similar experience will happen again. Two experiences is better, but it's still unclear what those experiences have in common. Three experiences allows the brain to rule out noise, and make reliable predictions about future events.
This is not to say that people can't learn from a single experience. We do that all the time. But as you have probably noticed, drawing conclusions from just one or two events is fraught with error. Very often the wrong lesson is learned.
This minimum requirement for new rules manifests in a variety of ways. For instance, in children's stories. Having three instances of something, be it goats, bears, or trips up a beanstalk, is a ubiquitous device, because it feels natural. The same is true in comedy.
As the excellent, and extremely funny Kate Offer once pointed out to me, you can't do the same thing in musical comedy three times. Do something funny once in a song, and the audience will laugh. Do it twice and the audience will love you for it--you've cued up their brains to think they know what's going on. Do it the same way a third time and the funny disappears. This is because by the time you've got to the end of learning experience 2, people are already projecting. To make the third iteration funny, you have to put a twist on your original gag which breaks expectations. If you do that, the audience will love you even more. You've shown them a meaningful pattern, but not the one they were expecting.
So if three iterations is so important, why am I proposing four steps for the perfect sketch? Doesn't 'subversion of pattern' count as step three? And how do we actually use these steps to craft good comedy? Any guesses?
I've probably said enough for one blog post, so I'll have to tell you next time. Meanwhile, all conjectures are welcome.
Tuesday, February 7, 2012
Promotion Dance
Want to know the secret to being promoted? Keen to ascend the ladder of success? Fancy seeing your hard work appreciated and reflected in a more senior position? Improv has the answer. Well, to be more exact, improv plus a healthy combination of agent-based modeling and behavior science has the answer. Or, at least, if not actually a complete answer, then at the very least a healthy dose of profound, if not slightly chilling insight.
Let me explain. One of the more compelling games that came out of Behavior Lab last year was something I call Promotion Dance. This was a game that I dreamt up during brainstorming sessions with the excellent Dave Sals. It grew out of a desire to create games that explained things about workplace dynamics, and also from some agent-based modeling work on age discrimination that I've been doing with Rich Martell, a fascinating and highly astute organizational psychologist I met recently.
To play this game you need a large group. Twenty people or so works well. These people each have one of three roles. CEO (you only have one of these), Team Lead (you need about three or four of these), and Team Members (everyone else). You arrange the people as follows:
Then, once everyone is positioned, you start dancing. The CEO's job is to communicate his dance choices to his team leads. The team leads' job is to communicate the CEO's choices to the teams. And the teams' job is to accurately reflect the dance they're seeing their team lead do. Team members are encouraged at the outset to focus on their team lead, not the CEO.
All that's left is to add the extra twist: that occasionally the facilitator will shout: Promotion! When that happens, the CEO has to choose the team lead he thinks is doing the best job. Then that team lead has to promote the member of their team who they in turn think is doing the best job. The selected team lead becomes the new CEO. The team member becomes the new team lead, and the CEO (this is where it gets unrealistic) fills the slot left by the promoted team member. The game then proceeds as before.
Promotion Dance is fun to play. People generally enjoy the dancing, and the sense of power that comes from being either a team lead or a CEO. However, it's what's revealed when a promotion happens that makes the game fascinating.
The first thing you notice is that without being asked, the CEO generally assumes that the best team lead is the one whose dance is most similar to their own, even though this has never been stated. The next fascinating thing is that the team leads have all been watching the CEO rather than their teams. This means that when one of them is picked, they suddenly remember that their teams exist and choose someone almost at random. What's even more fascinating is that these patterns persist, pretty much unaffected, even after people have started to figure out what's going on.
What do we learn from this?
Plenty. As always, we have to take the findings with a pinch of salt, because this is a game, not a real workplace. However, it's pretty clear that:
This demonstration is simplistic and undoubtedly we could muddy the waters with many different additions to the dynamic. However the point that the game makes when played is extremely clear. Without awareness and care, people automatically construct hierarchies that reward sycophancy. They don't do this because they're bad people, or have selfish personalities, or because their culture is toxic. It's simply the emergent effect of a hierarchical arrangement of people, all of whom want to receive attention.
It would be easy to say at this point that, sure, we know this happens. It's been going on for thousands of years. But that modern organizations use tools like assessment centers and 360 interviews to choose the people to promote. And while that's a great start it misses the core point. And that's that the recommendations for who to assess still come from somewhere, because HR departments can't watch everyone all the time. As we all know, tools for measuring the effectiveness of employees in their current roles won't really help you either, because you're assessing for talent in the position you're about to take your employee out of.
The easy, dangerous lesson we could absorb here is that kissing up to the boss really does help, for solid, testable reasons. The better lesson, I think, is that organizations that want to keep their culture healthy have to take active steps to avoid the promotion-dance-effect.
This starts with the CEO. Rather than looking just at his team leads because that's easy, he needs to be looking at the employees below to see what they're actually doing. However, it doesn't end there. If bottom-line employees want to be taken seriously, they need to make sure they're visible to people higher up the organization than their direct manager, and sharing clear information about their experiences whenever possible.
The right lesson, if you like, is that strong, healthy organizations dance together, rather than in a hierarchy.
Let me explain. One of the more compelling games that came out of Behavior Lab last year was something I call Promotion Dance. This was a game that I dreamt up during brainstorming sessions with the excellent Dave Sals. It grew out of a desire to create games that explained things about workplace dynamics, and also from some agent-based modeling work on age discrimination that I've been doing with Rich Martell, a fascinating and highly astute organizational psychologist I met recently.
To play this game you need a large group. Twenty people or so works well. These people each have one of three roles. CEO (you only have one of these), Team Lead (you need about three or four of these), and Team Members (everyone else). You arrange the people as follows:
Then, once everyone is positioned, you start dancing. The CEO's job is to communicate his dance choices to his team leads. The team leads' job is to communicate the CEO's choices to the teams. And the teams' job is to accurately reflect the dance they're seeing their team lead do. Team members are encouraged at the outset to focus on their team lead, not the CEO.
All that's left is to add the extra twist: that occasionally the facilitator will shout: Promotion! When that happens, the CEO has to choose the team lead he thinks is doing the best job. Then that team lead has to promote the member of their team who they in turn think is doing the best job. The selected team lead becomes the new CEO. The team member becomes the new team lead, and the CEO (this is where it gets unrealistic) fills the slot left by the promoted team member. The game then proceeds as before.
Promotion Dance is fun to play. People generally enjoy the dancing, and the sense of power that comes from being either a team lead or a CEO. However, it's what's revealed when a promotion happens that makes the game fascinating.
The first thing you notice is that without being asked, the CEO generally assumes that the best team lead is the one whose dance is most similar to their own, even though this has never been stated. The next fascinating thing is that the team leads have all been watching the CEO rather than their teams. This means that when one of them is picked, they suddenly remember that their teams exist and choose someone almost at random. What's even more fascinating is that these patterns persist, pretty much unaffected, even after people have started to figure out what's going on.
What do we learn from this?
Plenty. As always, we have to take the findings with a pinch of salt, because this is a game, not a real workplace. However, it's pretty clear that:
- When not given guidance to the contrary, people instinctively measure the quality of a person's work based on the attention they receive from that person, rather than from other measures that would require more work to evaluate. (This is because direct human attention is both rewarding and easy to assess. It fulfills our natural desire to have an effect on the world that is immediate, clear, and effortless to discern.)
- That when there are more than two levels in an organization, that the tension on those in the middle layers will always be resolved by those people directing more of their attention up the hierarchy than down. (This is because those who direct their attention down are less likely to get ahead, and will usually be replaced by those who direct more of their attention up.)
- That people who are below two levels of management must pay very close attention to their boss in order to be noticed, as the time they have to achieve this will be a small fraction of the time that their boss can possibly afford to give while remaining in the game.
This demonstration is simplistic and undoubtedly we could muddy the waters with many different additions to the dynamic. However the point that the game makes when played is extremely clear. Without awareness and care, people automatically construct hierarchies that reward sycophancy. They don't do this because they're bad people, or have selfish personalities, or because their culture is toxic. It's simply the emergent effect of a hierarchical arrangement of people, all of whom want to receive attention.
It would be easy to say at this point that, sure, we know this happens. It's been going on for thousands of years. But that modern organizations use tools like assessment centers and 360 interviews to choose the people to promote. And while that's a great start it misses the core point. And that's that the recommendations for who to assess still come from somewhere, because HR departments can't watch everyone all the time. As we all know, tools for measuring the effectiveness of employees in their current roles won't really help you either, because you're assessing for talent in the position you're about to take your employee out of.
The easy, dangerous lesson we could absorb here is that kissing up to the boss really does help, for solid, testable reasons. The better lesson, I think, is that organizations that want to keep their culture healthy have to take active steps to avoid the promotion-dance-effect.
This starts with the CEO. Rather than looking just at his team leads because that's easy, he needs to be looking at the employees below to see what they're actually doing. However, it doesn't end there. If bottom-line employees want to be taken seriously, they need to make sure they're visible to people higher up the organization than their direct manager, and sharing clear information about their experiences whenever possible.
The right lesson, if you like, is that strong, healthy organizations dance together, rather than in a hierarchy.
Friday, February 3, 2012
Archetypal Improv Revisited 2
In the name of clarity, fun, and general, all-purpose niceness, I hereby present some of the notes I recently shared with Opening Night Theater on the Vanilla Six Hander Format. The original question was: how do I teach it? My reply (now somewhat tidied up) was as follows:
- First, I make sure people have a strong grounding in status and listening skills.
- Then, I set expectations. Getting really good at full-length plays that might have been scripted usually takes about a year to really master. Also, some people find collaborating on complete plots a bit of a head-scrambler. People who just want to have fun on stage and don't enjoy having their brains burst often bow out before they absorb all the skills.
- I introduce the basic structure and why it's used.
- I run a workshop on getting into trouble. This often involves teaching tilts, escalation scenes, and some coached play openings.
- I teach just Scene 1 and give people plenty of opportunities to explore it until it's starting to come naturally. This sometimes comes with a small shock, because people are used to 'starting positive' to build a platform and avoiding 'instant trouble'. I encourage people to start working on dramatic tension from the first line if the story needs it by realizing that the best way to help a fellow improviser can often be to put them in an awkward situation that paints a sympathetic picture to the audience.
- I teach just Scene 2 without a scene 1 preceding it, so that people can get a sense of what it feels like in isolation.
- I build to Scene 1 followed by Scene 2, so that people can explore linking the two together and picking out themes that will work together.
- Once players have stabilized on how to link scenes together thematically while keeping the content distinct, I teach 1,2,3.
- I start getting people to ask the following questions. What is going to be the lowest moment for the protagonist? What is the best way for this story to end? I emphasize the fact that everyone's version of the obvious answers are going to be different, and that's okay. I also try to encourage people to actually be able to go dark in terms of plot content. Players, even very experienced ones, often find this very hard to do well. A lot of improv is taught with the mantra 'be positive and always say yes'. When you're doing long form, you often have to hurt a character to help a fellow improviser.
- I start coaching plays, letting them extrapolate forwards, usually encouraging players to drop a play at the point at which the troupe is feeling lost and start again. When a lot of scenes have been invested in a play, it's somewhat harder to think of it as 'disposable theater'.
- I coach players towards delivering a complete play, trying to help them see ways to get the protagonist in and out of trouble with escalation each time without making choices for them.
- When the troupe is able to pull off a complete play that they feel happy with, celebration is compulsory.
- Throughout the process, I watch the interpersonal dynamics very carefully. Constructing a full play is a challenge for many improvisers and requires developing new skills. That's often intensely rewarding for them but also tricky because some people pick it up faster than others. Also, some people inhabit their characters as they go, while others suddenly focus on plot and forget how to act. All this can cause friction that needs to be managed.
- I also make sure that I have at least one pencil and paper session in which I teach the troupe how to rapidly plot a movie or novel using the sequential approach. This provides a second lens through which to view the story-building process.
Monday, January 30, 2012
Archetypal Improv Revisited
A few weeks ago I learned that the lovely people at Opening Night Theater in Toronto wanted to explore the Vanilla Six Hander improv play format that I outlined on this blog back in April 2010. I was delighted.
I’ve been chatting with them via email since then, and thought it was about time to try to collect those thoughts, along with a few others. What I’m hoping to achieve in the next few posts is to give people a clearer sense of how best to explore the format, what its quirks are, and what kind of results it can deliver. First, some history.
The V6H grew out of two converging trains of thought that emerged out of the work I did with Amazing Spectacles in Cambridge around the turn of the millennium. The first of these was watching what happened when people tried to do unstructured improv plays. We were doing a lot of full-length plays at that point, and had various heuristics for how to build them. However, some clearly worked a lot better than others. Furthermore, the plays that worked had certain key features in common:
- A strong rapidly-defined platform
- Clear characters, but also clear character roles
- Threads that started somewhat separate but which joined into a single satisfying story arc
- Improvisers who were synced, in terms of their mood, skill levels, and expectations
From working with Patti Stiles in London, and from Freestye Rep in New York, I was familiar with both the idea of rising narrative tension, and Kenn Adam’s ‘story spine’ model. However, I had the sense that really good improv plays had certain symmetries that weren’t captured explicitly in either of these approaches.
Around the same time, Gary Mooney (one of the most naturally gifted comic improvisers I have ever met) pointed me at a book he had been reading called The Writer’s Journey, by Christopher Vogler. It was clear to me very quickly that the content in Vogler’s book had enormous implications for improv.
The Writer’s Journey is a tidy encapsulation of Joseph Campbell’s work on the monomyth from the writer’s perspective. Though it was intended for an audience of screenwriters rather than improvisers, it revealed patterns that occur in many of the great tales that have been told since the start of civilization--tales that were usually spoken rather than seen. In effect, it laid bare the mechanics of storytelling in any form.
For that reason, when it comes screenwriting, I take Vogler’s recipe with a pinch of salt, which I think he’d consider appropriate. This is because strict adherence to the monomyth pattern can produce stories that bear an uncanny resemblance to The Matrix or Star Wars, and which suffer a little in the subtlety department. You have to deviate from any recipe to make a story come alive, no matter how good that recipe is.
However, for improvisers, a strong, specific awareness of what makes a great story represents a massive advantage. This is because improvised stories always deviate from the recipe. The trick is to provide all your actors with a shared understanding of how to support each other and what direction to head in. Knowing how great stories are built, and how they’ve persisted for thousands of years, gives improvisers a golden compass to follow as they navigate the massive uncertainties of their art.
The problem we had was that the recipe that Vogler described was a pattern of sequential steps. In my experience, trying to hold an improv play to any specific sequence is tricky, and often problematic. The work that Kenn Adams has done seems to me to take this approach about as far as you reasonably can. Beyond that, you have too much structure in the work and the quality of the improv starts to deteriorate.
What my troupe in Cambridge needed was a parallelized version of the same approach. A model that created a platform so strong that the seeds of a great story arc were already latent within it. The V6H is still my best attempt to date to achieve that. It focuses on encouraging improvisers to take on archetypal roles that let the play feel rounded and purposeful while being truly improvised at the same time. The V6H works by structuring the first three scenes of the play fairly tightly and keeping the rest loose. However, once the principles of archetypal improv are embedded in actors’ minds, it’s safe to set down the structured introduction. By then, you usually know what a certain character is useful for about ten seconds after he or she walks on stage.
I have seen the V6H produce some really wonderful improvised theater. I’ve had the good fortune to watch as well as coach, and have seen it deliver some of the most funny, touching, shocking, dramatic, impressive long-format improv that I’ve ever witnessed. Furthermore, V6H plays have a very different feeling from those where the players simply rely on good listening and shared experience to guide them, even when the improvisers are very talented. Or from those plays that employ a platform-development-resolution kind of structure and don’t dig any deeper. That’s because V6H plays have a real shape. They breathe in and out like living things.
Having said that though, here some important points that anyone trying the V6H should bear in mind.
It’s not really mine
Though I have taken credit here for the V6H, the truth is that it was, and still is, a massive collective effort. Without Gary’s initial input, and without the significant refinements contributed by Dennis Howlett, Netta Shamir, Justin Lamb, and others, it’s unlikely that it would have ever got off the ground. The V6H then continued to evolve when I moved to Santa Cruz. More marvelous people, like Cindy Ventrice, Dave Sals, Tish Eastman, and Karen Menahan, helped me shape it. And the work is still going on in great troupes like Six Wheel Drive.
Because of this, there is no one set of rules for the V6H. There is no handbook (yet) and no notion of exact right and wrong. The format belongs to the people who are doing it. I think this is wonderful because it means it’s always evolving. However, its openness comes with difficulties because figuring out the perfect solution of how to use the V6H with your troupe is always something that will have to be figured out afresh.
The pain comes before the gains
Another key point is that building an understanding of plot and archetype into improv plays comes with costs before it comes with advantages. It can take people over a year to internalize the core principles. During that time, improvisers, often brilliant ones, will deliver a great number of wooden, clunky characters and terrible plot offers while they figure out how to integrate their new knowledge. This is because it's hard to serve the story and your character at the same time until you understand how the two sides fit together. It’s easy to feel like you’re going backwards during this time. Its only afterwards, when storytelling becomes this marvelous, collective, instinctive force, that you get all your original skills back.
There will be resistance
I have encountered a lot of improvisers (some of them really good) who don’t want to believe that great stories have a shape. I have encountered writers, playwrights, and screenwriters who believe this too. Even though many of these people tap into exactly the same patterns to make their work fly, they do so in a subconscious fashion without ever letting onto themselves what’s happening. The net is, if you’re running a troupe and people don’t want to believe that stories have structure, that’s a problem. It’s usually best under those circumstances to find a methodology that’s less transparent to work with, as cherished notions about story shape are seldom abandoned, with or without a fight.
There will be friction too
Developing a skill-set that requires that everyone pull together and collaborate on something complex creates friction. This is because everyone goes up the learning curve their own way and at their own speed. This can make small interpersonal difficulties that can show up in improv troupes suddenly seem to magnify and become horrible. It’s easy to believe that your little theater company is exploding, and sometimes they actually do. If you succeed, though, you’ll end up with a tightly-knit team of highly capable performers who can deliver work that’s as sharp as scripted theater.
Drills are your only real power-tools
Practicing frequently and relentlessly is the best way to develop V6H skills. Working on opening scenes until you can tell who the protagonist is going to be just by the way the first improviser walks on stage is going to make everything that follows that much easier. Drilling on nemesis scenes until you can assemble and propose an entire story premise in a single speech without once resorting to cliche takes time too. At the end of the day, good plays require a lot of non-declarative memory investment, and that takes time. Quite possibly, people will get bored at all the repeat play openings and complain. If that happens, do something else for a while. People only get better at improv when they’re motivated. Come back to the drills after you’ve tried a few more full-length plays and people have rediscovered how much they don’t know yet.
That's probably enough for now. Hopefully it's a useful start. For anyone out there playing with the V6H, I wish you the very best of luck. If you have any questions, I’m more than happy to address them. My your tension rise smoothly and your plays leave their audiences teetering, while laughing, on the edges of their seats.
Monday, June 13, 2011
Can Big (Companies) Be Beautiful?
As companies grow, they slow down. Once they get past a certain size, they usually experience what’s called organizational sclerosis--a culture change that makes effective adaptation almost impossible. They cling to old business models, become less responsive to their customers, and make big costly mistakes. Their shares drop.
People have various theories as to why this happens. Most of them relate to organizational complexity or conflict from within. The strategies for fixing it usually involve hiring expensive consultants or laying off hundreds of employees. These strategies are painful and often simply don’t fix the core problem. But what if there was a way to make large companies as dynamic as small ones? I think I might have found one.
Part of the Tokenomics research that I’ve been doing of late has involved building computer simulations of organizational cultures. (See previous post.) The simulations are still very simple and there’s lots of work still to be done before any of the results find their way into a science paper. However, some of the results I've discovered are pretty compelling. This is one of them.
I’ve been exploring the idea that people look for predictable, repeatable self-validation in their interactions with others, or ‘token collection’--an idea that you can find out more about on the SF Behavior Lab website, or in the in-depth documents stored at Techneq.com. Almost all the token collection that we do is driven by non-conscious pattern-seeking. This is very useful because it means that we don’t have to model complete human minds in order to get a rough picture of how cultures work. We can simply create populations of agents that interact by seeking out repeatable patterns of dialog, and modifying their behavior if their conversations don’t run as expected.
In the examples I described in my earlier post, the agents building cultures had a pretty limited set of options available to them and so the communities they formed didn’t show much variation. However, for the experiments I want to share with you this time, the interactions are rather different. Agents speak in turns, rather than at the same time, and have a choice of 128 different words they can say. (128 openings x 128 responses). This means that the agents have a much harder time of it finding familiar subjects to talk about with their peers, and so have to be a little creative (random) in their choices.
Most of the interactions that the agents try out don’t meet with much success. They produce fairly short-lived patterns shared only by a few robotic acquaintances. However, some interaction patterns manage to spread and dominate, just like fads in real life. Once these patterns get established, they become incredibly stable. They become ‘cultural interactions’ and end up being common to all the agents in the simulation. Removing or adding agents incrementally doesn’t change the culture one bit.
Also, interestingly, changing the size of the population of agents has a serious effect on how often new patterns get incorporated into the culture. Bigger populations have a lot more trouble accepting new ideas than small ones. At some level, though, this shouldn’t surprise us. In order for an idea to develop critical mass and spread rapidly through the group, a certain percentage of the agents have to have already heard it. That’s going to get steadily less likely as the system scales up.
For these simple agents, a critical drop-off happens before you get to a population size of forty. So far as I know, there’s no equivalent change in performance for human groups in this size range. The literature suggests that real organizations start having problems at around a hundred and fifty people. However, in the context of agent simulations, this makes sense. The agents have much smaller memories than people and thus less intrinsic flexibility. So, while the drop off in effectiveness may not be exactly the same phenomenon as organizational sclerosis, there seems to be a close enough fit here to make exploring the effect worthwhile.
The question I asked was ‘what do I have to do to a set of agents in order to keep their rate of cultural adaptation high while the population increases in size’. Specifically, what did I need to do to make a population of forty agents as imaginative as one of size twenty.
The answer, for agents, at least, is simple. You divide them into two groups of size twenty and let them run independently. Then after a while, you reshuffle them into two new groups, and leave them alone again. Every time the two groups are shuffled, learning from each community gets transferred into the total population. However, because the groups stay small, the rate of creativity stays at what you’d expect for a much smaller population.
This process is rather like the genetic recombination that happens in sexual selection. We keep the rate of change high by randomly mixing our ingredients and then making sure that any useful results get shared back into the population.
Also, it turns out that this isn’t the only method that will work. Any solution that involves keeping group sizes small most of the time, combined with random mixing, will do the job. And there appears to be an optimum group size for a given setting of the simulation parameters at which creativity can spread fastest.
The lesson seems clear: turning big organizations into a cross-polinating swarms of smaller ones might make them a lot more agile. While this approach probably isn’t a fit for all kinds of business, I can think of plenty where it might work: finance and software being the first two to spring to mind.
Once I’d managed to keep a big culture as adaptable as a small one, I wanted to know if I could do better. After all, large organizations have more people and therefore more ideas per day than small ones. In an ideal world, shouldn’t they be able to capitalize on that?
The answer is yes if you bend the rules a little. I couldn’t find a way of reorganizing the teams that led to further improvements in creativity beyond what I’d already tried. However, when I started messing with the parameters of the simulation, I noticed something interesting: increasing the number of possible witnesses to each interaction by one caused the rate of creativity to jump up rapidly. One way to say this is that by raising the number of people in a team who get to listen in on important interactions increases the rate at which new ideas can spread
This provides some simulation evidence that the people in the Agile Software Development movement are on the right track. They’ve been saying for years that creating small, tightly-knit teams where information is regularly shared creates the right environment for creative work. Furthermore, the agents in this simulation are so simple and general that we can expect the same logic to hold for a large number of possible situations.
The important next step will be to see whether the same tricks work as I add in more of the complexity of real organizations. For instance, trust relationships, hierarchies, and differences in effective value between the habits being propagated all have yet to be added. If the splitting/merging effect still works when all these features are added in, it might be time to start rethinking how we grow companies. Organized packs of collaborative mammals could turn out to be a lot more adaptable than a few big dinosaurs.
People have various theories as to why this happens. Most of them relate to organizational complexity or conflict from within. The strategies for fixing it usually involve hiring expensive consultants or laying off hundreds of employees. These strategies are painful and often simply don’t fix the core problem. But what if there was a way to make large companies as dynamic as small ones? I think I might have found one.
Part of the Tokenomics research that I’ve been doing of late has involved building computer simulations of organizational cultures. (See previous post.) The simulations are still very simple and there’s lots of work still to be done before any of the results find their way into a science paper. However, some of the results I've discovered are pretty compelling. This is one of them.
I’ve been exploring the idea that people look for predictable, repeatable self-validation in their interactions with others, or ‘token collection’--an idea that you can find out more about on the SF Behavior Lab website, or in the in-depth documents stored at Techneq.com. Almost all the token collection that we do is driven by non-conscious pattern-seeking. This is very useful because it means that we don’t have to model complete human minds in order to get a rough picture of how cultures work. We can simply create populations of agents that interact by seeking out repeatable patterns of dialog, and modifying their behavior if their conversations don’t run as expected.
In the examples I described in my earlier post, the agents building cultures had a pretty limited set of options available to them and so the communities they formed didn’t show much variation. However, for the experiments I want to share with you this time, the interactions are rather different. Agents speak in turns, rather than at the same time, and have a choice of 128 different words they can say. (128 openings x 128 responses). This means that the agents have a much harder time of it finding familiar subjects to talk about with their peers, and so have to be a little creative (random) in their choices.
Most of the interactions that the agents try out don’t meet with much success. They produce fairly short-lived patterns shared only by a few robotic acquaintances. However, some interaction patterns manage to spread and dominate, just like fads in real life. Once these patterns get established, they become incredibly stable. They become ‘cultural interactions’ and end up being common to all the agents in the simulation. Removing or adding agents incrementally doesn’t change the culture one bit.
Also, interestingly, changing the size of the population of agents has a serious effect on how often new patterns get incorporated into the culture. Bigger populations have a lot more trouble accepting new ideas than small ones. At some level, though, this shouldn’t surprise us. In order for an idea to develop critical mass and spread rapidly through the group, a certain percentage of the agents have to have already heard it. That’s going to get steadily less likely as the system scales up.
For these simple agents, a critical drop-off happens before you get to a population size of forty. So far as I know, there’s no equivalent change in performance for human groups in this size range. The literature suggests that real organizations start having problems at around a hundred and fifty people. However, in the context of agent simulations, this makes sense. The agents have much smaller memories than people and thus less intrinsic flexibility. So, while the drop off in effectiveness may not be exactly the same phenomenon as organizational sclerosis, there seems to be a close enough fit here to make exploring the effect worthwhile.
The question I asked was ‘what do I have to do to a set of agents in order to keep their rate of cultural adaptation high while the population increases in size’. Specifically, what did I need to do to make a population of forty agents as imaginative as one of size twenty.
The answer, for agents, at least, is simple. You divide them into two groups of size twenty and let them run independently. Then after a while, you reshuffle them into two new groups, and leave them alone again. Every time the two groups are shuffled, learning from each community gets transferred into the total population. However, because the groups stay small, the rate of creativity stays at what you’d expect for a much smaller population.
This process is rather like the genetic recombination that happens in sexual selection. We keep the rate of change high by randomly mixing our ingredients and then making sure that any useful results get shared back into the population.
Also, it turns out that this isn’t the only method that will work. Any solution that involves keeping group sizes small most of the time, combined with random mixing, will do the job. And there appears to be an optimum group size for a given setting of the simulation parameters at which creativity can spread fastest.
The lesson seems clear: turning big organizations into a cross-polinating swarms of smaller ones might make them a lot more agile. While this approach probably isn’t a fit for all kinds of business, I can think of plenty where it might work: finance and software being the first two to spring to mind.
Once I’d managed to keep a big culture as adaptable as a small one, I wanted to know if I could do better. After all, large organizations have more people and therefore more ideas per day than small ones. In an ideal world, shouldn’t they be able to capitalize on that?
The answer is yes if you bend the rules a little. I couldn’t find a way of reorganizing the teams that led to further improvements in creativity beyond what I’d already tried. However, when I started messing with the parameters of the simulation, I noticed something interesting: increasing the number of possible witnesses to each interaction by one caused the rate of creativity to jump up rapidly. One way to say this is that by raising the number of people in a team who get to listen in on important interactions increases the rate at which new ideas can spread
This provides some simulation evidence that the people in the Agile Software Development movement are on the right track. They’ve been saying for years that creating small, tightly-knit teams where information is regularly shared creates the right environment for creative work. Furthermore, the agents in this simulation are so simple and general that we can expect the same logic to hold for a large number of possible situations.
The important next step will be to see whether the same tricks work as I add in more of the complexity of real organizations. For instance, trust relationships, hierarchies, and differences in effective value between the habits being propagated all have yet to be added. If the splitting/merging effect still works when all these features are added in, it might be time to start rethinking how we grow companies. Organized packs of collaborative mammals could turn out to be a lot more adaptable than a few big dinosaurs.
Wednesday, June 8, 2011
Adventures in Game Theory, Part Four
For those of you freshly joining this adventure, the last three posts have led us on a strange, thrilling journey that has passed through the valleys of introductory game theory, the jungles of applied improv, and the mountains of software simulation. Now, at last we arrive at our thunderous finale on the shores of Lake Awesome. I highly recommend reading from the start of the sequence, otherwise what I have to say may be too extraordinary and wonderful for your mind to fully hold!
At the end of the last installment caught me teetering on the brink of a realization--that by adding just a little more functionality to my simulation, I could start exploring some more socially useful truths about how people behave. My insight was to add status.
What this meant in practice was splitting the population of agents in my model into two groups: bosses and workers, or in training community parlance: leaders and team-members. Then, in order to make the interactions between bosses and workers a little less benign, I added two extra constraints.
One: If bosses were aggressive (nose-thumbing) to workers, workers were not empowered to reciprocate and be aggressive back in their next encounter.
Two: Bosses were unable to remember the specifics of positive interactions they had with workers. So for instance, if a boss and a worker both chose paper in one round, the worker would remember the fact, but the boss would not.
Implementing these changes was easy, as it simply required that the two memory rules I’d already added to make the first simulation work were now dependent on status. (I also added a little extra logic around the movement of the agents to ensure that workers had to interact with bosses, and to make the movements of bosses dependent on other bosses but not workers. However, while necessary, that code is somewhat beside the point.)
What happened next was wonderfully clear. Within a few seconds, all the bosses were behaving aggressively while the workers normed on a set of social standards of their own. My simulation suddenly looked a lot like some of the more awful companies I’d worked for. Without having to say anything about the kinds of people who become leaders, or about the specifics of organizational culture, I’d captured a simple truth about leadership: that without the incentives to behave otherwise and the right skills to succeed, people with power slide towards bad behavior, even if they start off thinking like saints.
What was even more interesting was that as the simulation progressed, the bosses started to bump up against the corners of the virtual environment as if desperate to leave. Because aggressive behavior was so successful for bosses in their interactions with workers, they were applying the same behavior to each other, resulting in a rapid erosion of their ability to collaborate. The lesson: by letting leaders behave badly, we ensure that leaders have less pleasant interactions with each other, as well as with us.
My goal, though, was not to engage in rhetoric about leaders, but instead to see whether models like the one I was looking at could tell us something about how to help organizations do better. To do this, I looked at what happened when I turned each of the status dependencies off in isolation.
Turning off the status dependency for remembering positive interactions is rather like sending your managers on an employee recognition course. They learn to value the specific information they get from each person they work with, and to let their team members know that they’re seen and valued.
The result in the simulation is that the culture improves significantly. The workers integrate more tightly and the bosses take on the same cultural colors as the workers they lead. Interestingly, the bosses don’t all start cooperating at once. Many of them initially retain their aggressive behavior. Then, one by one, they figure out that collaboration is more effective.
The lesson here: that training leaders to listen can make a huge difference in their effectiveness, but that the change they take on depends on their willingness to implement what they learn.
If instead, we turn off the status dependency for worker retaliation to boss aggression, the effects are even more interesting. Making this change is rather like implementing a shared accountability system like the one that revolutionized the airline industry and transformed the safety standards in air travel. Under this system, the pilots of planes are no longer the unquestionable captains of the air that they once were. If copilots think that they’re witnessing a mistake, they’re duty-bound to call the pilot on it and to report it to air traffic control if necessary. In our simulated business, we can imagine that we’re instructing the worker agents to hold their bosses accountable if they don’t uphold the collaborative social standards of their organization.
What happens when we make this change is that the behaviors of the bosses have trouble settling onto any specific color. When we watch the ‘mood’ of the agents to see how many positive or negative interactions they’re having, we see that the tables have been turned. The workers are now having a pretty great time all round and the bosses are mostly miserable--the opposite of what we see if status dependence for retaliation is left on. This is because the workers now have an advantage that the bosses don’t--they can remember and repeat positive interactions whereas bosses cannot. Because aggression no longer secures automatic results, bosses don’t have an easy way of stabilizing on a successful behavior.
The lesson here is that enabling everyone in an organization to hold leaders accountable for their behavior is what creates the incentive for leaders to improve, but that without the right training and direction, the main result is leader unhappiness.
As you might expect, turning off both status-dependent features creates a benign, functional organization that settles rapidly onto a cooperative culture. If you want to play around yourself, and have Java installed, the simulation is the second applet on this page. (It has four buttons.)
As before, red, blue and green denote different positive interactions. Gray denotes aggressive behavior. Swapping to ‘mood view’ shows the success of the agents interactions, ranging from blue (unhappy agents) to yellow (cheerful ones).
Clearly there’s a lot more to do here. For a start, in order to turn this into a science result, the simulations will need to be a lot more rigorous, which will probably mean sacrificing the visual playfulness. Furthermore, we’ve only looked at one memory model for agents and solid research would need to try out others. However, the results seem pretty clear. We’ve gone from a simple game played in a room full of people to a model that turns business intuition into something rather like unavoidable, mathematical fact.
Thus, in the wake of our adventure, we can say with real confidence that any society or organization that doesn’t empower its people hold its leaders accountable, and which doesn’t teach those leaders how to listen, can expect its leaders to turn out bad, regardless of how ‘good’ we believe them to be as people.
This is something most of us already believe but which we often fail to implement. For instance, we're all used to the idea of holding elected officials accountable, but explicit training in 'voter recognition'? We leave that to chance. Similarly, we're used to the idea that good managers are the ones who pay attention, but company-wide accountability systems? Those are pretty rare. I believe that simulations like this can make these points unavoidable, and also perhaps show us how to build measures that make our adherence to such standards quantifiable.
For any skeptics out there, my huge thanks for reading this far, and here’s a final thought to consider. Agent-based simulations of this sort have been used by biologists for years on the following basis: we can’t capture all the details of natural systems like cultures or the lives of organisms, so instead we capture only what we know is true. From that, we look to see what else must be true as a consequence. Thus we attempt to make the simplicity of the model a strength, not a weakness. In this instance, the agents are so simple that we can expect the same effects to arise regardless of the memory model we employ for our agents, so long as that memory model permits learning. Further work in this area will hopefully make that point even clearer.
That’s it. The adventure is finished. And while the ending perhaps isn’t unexpected, it feels like a step forwards to me. After all, if we can do this starting with Rock Paper Scissors, think what we can do with the game of Twister.
At the end of the last installment caught me teetering on the brink of a realization--that by adding just a little more functionality to my simulation, I could start exploring some more socially useful truths about how people behave. My insight was to add status.
What this meant in practice was splitting the population of agents in my model into two groups: bosses and workers, or in training community parlance: leaders and team-members. Then, in order to make the interactions between bosses and workers a little less benign, I added two extra constraints.
One: If bosses were aggressive (nose-thumbing) to workers, workers were not empowered to reciprocate and be aggressive back in their next encounter.
Two: Bosses were unable to remember the specifics of positive interactions they had with workers. So for instance, if a boss and a worker both chose paper in one round, the worker would remember the fact, but the boss would not.
Implementing these changes was easy, as it simply required that the two memory rules I’d already added to make the first simulation work were now dependent on status. (I also added a little extra logic around the movement of the agents to ensure that workers had to interact with bosses, and to make the movements of bosses dependent on other bosses but not workers. However, while necessary, that code is somewhat beside the point.)
What happened next was wonderfully clear. Within a few seconds, all the bosses were behaving aggressively while the workers normed on a set of social standards of their own. My simulation suddenly looked a lot like some of the more awful companies I’d worked for. Without having to say anything about the kinds of people who become leaders, or about the specifics of organizational culture, I’d captured a simple truth about leadership: that without the incentives to behave otherwise and the right skills to succeed, people with power slide towards bad behavior, even if they start off thinking like saints.
What was even more interesting was that as the simulation progressed, the bosses started to bump up against the corners of the virtual environment as if desperate to leave. Because aggressive behavior was so successful for bosses in their interactions with workers, they were applying the same behavior to each other, resulting in a rapid erosion of their ability to collaborate. The lesson: by letting leaders behave badly, we ensure that leaders have less pleasant interactions with each other, as well as with us.
My goal, though, was not to engage in rhetoric about leaders, but instead to see whether models like the one I was looking at could tell us something about how to help organizations do better. To do this, I looked at what happened when I turned each of the status dependencies off in isolation.
Turning off the status dependency for remembering positive interactions is rather like sending your managers on an employee recognition course. They learn to value the specific information they get from each person they work with, and to let their team members know that they’re seen and valued.
The result in the simulation is that the culture improves significantly. The workers integrate more tightly and the bosses take on the same cultural colors as the workers they lead. Interestingly, the bosses don’t all start cooperating at once. Many of them initially retain their aggressive behavior. Then, one by one, they figure out that collaboration is more effective.
The lesson here: that training leaders to listen can make a huge difference in their effectiveness, but that the change they take on depends on their willingness to implement what they learn.
If instead, we turn off the status dependency for worker retaliation to boss aggression, the effects are even more interesting. Making this change is rather like implementing a shared accountability system like the one that revolutionized the airline industry and transformed the safety standards in air travel. Under this system, the pilots of planes are no longer the unquestionable captains of the air that they once were. If copilots think that they’re witnessing a mistake, they’re duty-bound to call the pilot on it and to report it to air traffic control if necessary. In our simulated business, we can imagine that we’re instructing the worker agents to hold their bosses accountable if they don’t uphold the collaborative social standards of their organization.
What happens when we make this change is that the behaviors of the bosses have trouble settling onto any specific color. When we watch the ‘mood’ of the agents to see how many positive or negative interactions they’re having, we see that the tables have been turned. The workers are now having a pretty great time all round and the bosses are mostly miserable--the opposite of what we see if status dependence for retaliation is left on. This is because the workers now have an advantage that the bosses don’t--they can remember and repeat positive interactions whereas bosses cannot. Because aggression no longer secures automatic results, bosses don’t have an easy way of stabilizing on a successful behavior.
The lesson here is that enabling everyone in an organization to hold leaders accountable for their behavior is what creates the incentive for leaders to improve, but that without the right training and direction, the main result is leader unhappiness.
As you might expect, turning off both status-dependent features creates a benign, functional organization that settles rapidly onto a cooperative culture. If you want to play around yourself, and have Java installed, the simulation is the second applet on this page. (It has four buttons.)
As before, red, blue and green denote different positive interactions. Gray denotes aggressive behavior. Swapping to ‘mood view’ shows the success of the agents interactions, ranging from blue (unhappy agents) to yellow (cheerful ones).
Clearly there’s a lot more to do here. For a start, in order to turn this into a science result, the simulations will need to be a lot more rigorous, which will probably mean sacrificing the visual playfulness. Furthermore, we’ve only looked at one memory model for agents and solid research would need to try out others. However, the results seem pretty clear. We’ve gone from a simple game played in a room full of people to a model that turns business intuition into something rather like unavoidable, mathematical fact.
Thus, in the wake of our adventure, we can say with real confidence that any society or organization that doesn’t empower its people hold its leaders accountable, and which doesn’t teach those leaders how to listen, can expect its leaders to turn out bad, regardless of how ‘good’ we believe them to be as people.
This is something most of us already believe but which we often fail to implement. For instance, we're all used to the idea of holding elected officials accountable, but explicit training in 'voter recognition'? We leave that to chance. Similarly, we're used to the idea that good managers are the ones who pay attention, but company-wide accountability systems? Those are pretty rare. I believe that simulations like this can make these points unavoidable, and also perhaps show us how to build measures that make our adherence to such standards quantifiable.
For any skeptics out there, my huge thanks for reading this far, and here’s a final thought to consider. Agent-based simulations of this sort have been used by biologists for years on the following basis: we can’t capture all the details of natural systems like cultures or the lives of organisms, so instead we capture only what we know is true. From that, we look to see what else must be true as a consequence. Thus we attempt to make the simplicity of the model a strength, not a weakness. In this instance, the agents are so simple that we can expect the same effects to arise regardless of the memory model we employ for our agents, so long as that memory model permits learning. Further work in this area will hopefully make that point even clearer.
That’s it. The adventure is finished. And while the ending perhaps isn’t unexpected, it feels like a step forwards to me. After all, if we can do this starting with Rock Paper Scissors, think what we can do with the game of Twister.
Subscribe to:
Posts (Atom)