Monday, February 7, 2011

The Science Incubator Game

Science is fracturing. People from different fields don't really understand each other's work all that well. Specialty areas keep getting smaller and more focused. Furthermore, many scientists have to operate in a culture that discourages people from opening their mouths if they don't understand what's being said. This is because how 'brilliant' others imagine you to be often has immediate repercussions for the job you get next. This unwillingness to speak up only makes the fracturing happen faster.

I can't see this culture of caution ending any time soon without outside help because it's driven by two things:
  1. It's just harder to understand what people in other fields are doing these days because the amount of understanding that you have to invest to reach the coal-face of science is hugely more than it used to be. Consequently, people try less. 
  2. The pressure in the scientific job market is incredible and it's getting worse. Gone are the days when people walked straight from their PhDs to faculty jobs. The incentives for people to open their mouths and risk looking foolish have never been lower. 
This whole trend is unfortunate, because the research shows that interdisciplinary dialog accelerates progress. Groups with mixed skill sets consistently find solutions faster than teams of people who specialize in the same subfield. The act of having to articulate your ideas to those who may not understand is not only going to force you to bring order to your own ideas, but is also likely to lead to the creation of new ones. Creativity, as it turns out, is not driven by sudden sparks of spontaneous genius, but by a process of blending pre-existing notions. Whether this happens inside a single person's head or in a social context doesn't seem to matter.

Furthermore, social innovation is always best activated by play and many of today's scientific workplaces are still lamentably low on playfullness. Whereas software companies have incorporated all manner of tools for establishing a sense of fun into their offices, many scientific departments still imagine that it's somehow 'more professional' to have people sitting in silence in small offices.

So, is there something we can do to fix this? Can we use applied improv to make science healthier, smarter, and more playful? I think so, and here's my best guess so far as to how to do it.

The trick is to play the science incubator game. For this you need:
  • A cafe.
  • A volunteer to be master of ceremonies.
  • About six people who like to learn and think. 
The science incubator game takes the form of an open, flowing dialog. An ideal session is likely to last from one to two hours, depending on everyone's stamina, and works like this.
  • One person, the 'proposer' brings along an open question that they're trying to answer. This can be as abstruse and as deep into their work as they like. In fact, the more abstruse, the better. The proposer tries to explain their problem to the rest of the group.
  • The others ask questions every time what they're hearing isn't clear, and chuck in any ideas that seem relevant. Everyone else in the group to tries to learn, and be as supportive as possible.
  • The master of ceremonies acts as an adjudicator, making sure that everyone gets a voice and that the rules are followed.
  • The session should start with everyone in the group telling a deliberate lie, so as to activate the creative parts of their brains. The more confusing or elaborate the lies are, the better. 
The rules of the game are:
  • All questions and ideas are good. Nobody gets to pass judgement on anyone's question or idea, regardless of how flawed they think it is. The proposer should try to answer all questions that are asked. (Why: Uses the Yes-And principle to create a shared narrative.)
  • Interruption should be done politely, but is mandatory. If the dialog has gone on for five minutes without somebody chiming in with a thought or suggestion, the master of ceremonies asks a question of his own. (Why: Prevents grandstanding and encourages group ownership of the process.)
  • Silence is banned. If a silence lasts for more than five seconds, the master of ceremonies should chime in with a new question. (Why: To maintain the energy level.)
  • Everyone is equal. All work hierarchy is left at the door when the incubator game is in progress. Anyone who pulls rank, or attempts to refer to their depth of experience to validate a point gets an immediate reprimand from the master of ceremonies. (Why: To help the space feel safe to all, and removed from normal patterns of social cost.)
  • Negativity is banned. In academic settings, people often consider their value to be in filtering out the proposals that won't work. In the incubator game, this role is forbidden. The way to add value is to add more ideas. (Why: Prevents contributors from self-awarding value via 'critical rationality' and derailing the session at the same time.)
  • Everything is informal. The purpose of the incubator is to reduce the risks of making suggestions and asking questions. Everyone is there to learn. Humor is strongly encouraged. (Why: Laughter activates the signal for social learning.)
  • Everyone must contribute. If someone has been quiet, or has been left out of the process, it's the master of ceremonies's job to bring them back in and make sure they feel safe. (Why: To encourage acceptance of the 'price of entry' of the session, which is engagement.)
  • Everyone should try to insert at least one harebrained suggestion that they have just thought of without considering the implications properly. Anyone who confesses 'I have no idea what I'm talking about' should get an immediate cheer. (Why: To break the dangerous social habit of over-filtering ideas out of perceived risk.)
  • The dialog stays on topic until the session is over, and shouldn't deviate into gossip. (Why: To break the idea that 'shop talk' is somehow dull, and to create as much engagement in new ideas as possible.)
  • Nobody is 'on show'. If the master of ceremonies feels that the game is dissolving into a performance of sorts between a few people in the group, or if people are waiting to have a 'good idea' before chiming in, the master of ceremonies needs to fix the balance. If necessary, the master of ceremonies can ask a specific person in the group for 'a half-baked idea, please'. (Why: To prevent social dominance patterns from forming within the game.)
  • This game doesn't have to be played in a cafe, of course, but ground that feels neutral and safe is a good idea. At someone's house over a shared pizza would work equally well. A glass of wine might also assist the process. 
  • Also, it clearly doesn't need to be done by scientists, either. I suspect that any group of people who're looking to share ideas and brainstorm would probably benefit from something like this. 
  • Leaving out rules that aren't working, and letting the proces flex to reflect the needs of the group is a good idea, so long as the spirit of the rules is maintained. 
Ways to make the process stronger:
  • If everyone brings pencil and paper, and tries building pictures or mind-maps of what they're hearing while others are speaking, it's likely to broaden thinking and keep the energy level up. (Why: Using the visual parts of the brain is likely to increase engagement.) 
  • Courageous, committed scientists should wear unusual hats, or pin large, ridiculous flowers to their lapels, or some such thing. Any symbol that creates a sense of group inclusion and of willingness to be silly is commended. (Why: Anything which fosters laughter is likely to improve the quality of social learning, and the bond that's formed.) 
  • If a proposer focuses on recent work they've done which went badly wrong, and talks over what happened with positive input from others, this is likely to help boost the proposer's learning experience significantly. (Why: Research shows that negative reinforcement without emotional consequences maximizes learning rates.)
  • Wind down time after the game is probably a good idea. If people get to keep talking and laughing after the end of the session, it's likely to improve the sense of group connection. (Why: The game is likely to encourage intensity. People will feel the need to re-establish working norms afterward.)
I've no idea whether this game works or not because I haven't tried it yet. To some extent, this game simply outsources the difficulties of scientific collaboration to the master of ceremonies--a kind of magical mediator who requires the courage and wisdom to step in every time the dialog falls off track. However, finding such a mediator might not be so hard. The role that I've outlined here is something that every decent improv instructor knows how to do. The necessary skill sets are surely easy to find if we go looking for them in our communities or take the time to acquire them for ourselves.

I'll be running some experiments here in Berkeley if I can find enough brave scientists. If anyone has any suggestions on the format, or gets to try it out before I do, I'd be delighted to hear about it.

Tuesday, February 1, 2011

TED Talks I Like

I like TED talks. They’re a marvelous way of getting access to many fascinating ideas in a very short time. (They’re also a fascinating series of examples of what does and does not make for compelling public speaking, but that’s a whole other blog entry.) At the suggestion of my good friend at SFBehaviorLab, David Sals, I’ve put together a list of some of the talks I like.

First, here are some talks by authors I’ve already raved about in my recent reading list post. These people therefore need no introduction. 

And here are are a few that I’ve recently encountered, all of which I think deserve a viewing. 

This has to be a great place to start. Clearer evidence of the profound neurological effect of improvisation would be hard to find. 

This is a great one to watch after the Dan Ariely talk listed above. Laurie Santos clearly demonstrates that the kind of decision-making ‘mistakes’ we make aren’t specific to the human race. This suggests that these patterns of reasoning are old. To my mind, this has important implications.
The patterns of decision-making that behavioral economics has revealed don’t just tell us things about how people react. They’re very likely to be providing us with important insights about how effective reasoning works. These ‘mistakes’ have been selected for over the course of millions of years of evolution. If they cause us to make some choices ineffectively, there must be other advantages that we gain. Though it may not be clear yet exactly what those gains are, experiments in Machine Learning are likely to help us find out. 

What I like about this talk is the distinction that Prof. Kahneman makes between the way we experience things and the way we remember them. He points out that the connection between the two is far shakier than we’d like to imagine. For me, this says interesting things both about the nature of declarative memory, and how we can use it to make our interactions with each other better. For instance, it seems clear that following negative feedback to a person with something a little nicer is likely to cause that person to walk away with a far rosier impression of the experience than if only negative input is received. This suggests that fixing some toxic workplace interactions may be as simple as bolting positive rituals onto the end of them--a fascinating implication, if it’s right. 

This talk is on how we can use an understanding of social networks to gain insights about the spread of diseases, social trends, and even emotions. Most significantly, Prof. Christakis reveals a simple mechanism by which we can identify ‘hubs’ in social networks and use them to gain advance warning of changes sweeping through a population. However, he also shows that interacting with these hubs provides us with a way to intervene as well as to watch. For instance, it tells us how to best to deploy a vaccine into a population to save the maximum number of lives.
The implication for applied improv here is that the same tools enable us to find those members of a community most likely to A: reflect the values of a culture, and B: change them, if we can engage them and give them the right tools. 

Mr. Johnson’s research into the the kind of social environments that foster good ideas feel like a natural fit for applied improv. Lurking in here, I feel, are clues as to how to use the science of play and the study of behavioral games to create innovation incubators. This talk leads me to wonder what kinds of improv you can play sitting down with four molecular biologists in a Starbucks without having anyone raise their voice or leave their seat. My suspicion is that one can do quite a lot, and probably get some publication worthy material out of it at the same time.