Why brainstorming doesn’t work

edited November 2011 in Stuff to Watch
Why brainstorming doesn’t work...
Evidence has long shown that getting a group of people to think individually about solutions, and then combining their ideas, can be more productive than getting them to think as a group. Some people are afraid of introducing radical ideas in front of a group and don’t speak up; in other cases, the group is either too small or too big to be effective.
But according to a recently published study, the real problem may be that participants’ get stuck on each others’ ideas.
Kohn and Smith believed the cause might be due to “cognitive fixation,” or the concept that, when exposed to group members’ ideas, people focused on those and blocked other types of ideas from taking hold. They experimented with this by manipulating the number of ideas participants saw in their chat windows, with some getting a few cues and others getting more. Their hypothesis was right: When exposed to many cues, the undergrads offered up less creative, diverse ideas. The numbers improved when the students were given a five-minute break during the exercise.
As with all such studies, there’s plenty of pretty obvious common sense to this research. But it’s a helpful reminder of how unhelpful it can be when managers dump people in a room together, thinking it will result in creative big ideas. Somehow, a belief in the power of group brainstorming sessions persists, despite evidence that it doesn’t work.
Possibly relevant to RPGs?

Comments

  • Absolutely relevant to RPGs! I think that how people decide what to go with is important, though -- in my experience, consensus-building tends to drag out discussion and give people an excuse to become incredibly fiddly with every last little word and phrase, whereas an up-down yes/no vote tends to force people to accept the overall spirit of an idea, opine upon it, and then vote & move on.
  • edited November 2011
    Of course, it probably depends how the brainstorming is run. I've found that for this kind of thing it's useful to have a group leader who encourages participation and creativity, as well as time for thinking on your own. Hurrah for GMs!

    Also, less ideas might just mean that the stupid ones don't get mentioned. It's about quality not just quantity.
  • I run into this all the time. I thrive on collaboration but once you introduce a larger group of people (and, perhaps, playtesters) it can start to produce roadblocks to creative solutions. That said, this time around I hired a team of concept people to throw ideas at me so that I could get to writing the game I wanted to write, and I've also brought in the feedback of others, and while it's at times incredibly frustrating and you need to schedule everyone together it's also been the firehose of awesome that I hoped.

    Cheers,
    Cam
  • This is why I like Microscope a lot.
  • I'm immediately struck at how you could replace the term "brainstorming" with the term "groupthink".
  • I think the key is to have everyone bring stuff to the table and then discuss. For example, if you want your first session to be world building and character generation then give everyone a simple piece of homework like writing a top 5 list of setting details or a relationship that they want to explore. When they get there, everyone can pool their lists and go over it. On the other hand, if people show up and then you start asking people how they want the game to go, then you probably will see some groupthink and/or subsequent people spending more time riffing off previous people's ideas rather than introducing their own.
  • There are lots of different techniques for brainstorming that recognise some of the problems and attempt to get round them. I've had a fair amount of training in this as part of operational research (OR). It's part of Soft OR (more people focused) as opposed to Hard OR (more about the numbers).

    It's always been interesting to me to note the parallels between a gaming session and a problem solving session and the extent to which metrics for one can be ported to the other. It's also quite fun, as a GM, to do some SWOT analysis for someone's character. It ends up looking very similar to something that's in the Dresden Files. Or alternatively, try creating a root definition using the CATWOE part of Soft-Systems Methodology as a GM to think about where your game might go next.
  • I've never cared for group brainstorming--I seriously identify the word more with when I think up a list of 20 things and use 10 of them--but that's partly me being an introvert and tending to prefer to think about things in a slow, deliberate way. Plus for my own projects I tend to have a very strong personal vision, and by the time I feel I need help a typical unstructured brainstorming session tends to bring in ideas I'm just not interested in.

    In RPGs I find that blank, unstructured creativity is incredibly hard, and even a tiny bit of structure can make it vastly easier and more effective. The Pathways system in Smallville is hugely helpful, and for a Wushu campaign we just went around the table letting people take turns adding ideas to the setting, which was far better than just doing it informally (we also tried out my Pathways-inspired Entanglements relationship mapping thing, which needs some work, but was very helpful).
  • edited November 2011
    But we're not brainstorming in RPGs. We're building on each other's ideas, rather than trying to generate as many as possible. So, presumably, focussing on other people's ideas ("cognitive fixation") is a benefit.

    (Interesting note. American psychologists generally suggest that groups are bad: c.f. groupthink. British psychologists generally suggest that groups are good: c.f. Social Identity Theory. Keep sceptical.)
  • And American OR doesn't recognise Soft OR at all. Even with brainstorming there comes a point where you have to choose between the options you've generated and there are several ways of doing that too, such as Multicriteria Decision Analysis or Strategic Choice Approach.
  • This is one reason why being a loudmouth Producer who says some ideas are shit is the best way to handle series creation in PTA.
  • Posted By: GB SteveAnd American OR doesn't recognise Soft OR at all.
    Soft OR?
  • Posted By: GrahamBut we're not brainstorming in RPGs. We're building on each other's ideas, rather than trying to generate as many as possible. So, presumably, focussing on other people's ideas ("cognitive fixation") is a benefit.

    (Interesting note. American psychologists generally suggest that groups arebad: c.f. groupthink. British psychologists generally suggest that groups aregood: c.f. Social Identity Theory. Keep sceptical.)
    I think the OP is referring to some games that have 'brainstorming' specfically as a part of play, such as - collectively outline an initial situation or something. Like in PTA , the first thing everybody does is agree on what type of TV series the game is going to be about. The alternative being someone hands down the ideas from on high (the gM?) or people come to the table with their own fully fleshed ideas and the group collectiovely choose one.
  • Posted By: stefoidI think the OP is referring to some games that have 'brainstorming' specfically as a part of play, such as -collectively outline an initial situationor something. Like in PTA , the first thing everybody does is agree on what type of TV series the game is going to be about.
    Yes, but that's not brainstorming. Brainstorming is, specifically, generating as many ideas as possible as a group. The evidence is that it's better to generate ideas individually, then bring them to a group.

    But that doesn't sound relevant to RPGs.
  • edited November 2011
    Posted By: GrahamBut we're not brainstorming in RPGs. We're building on each other's ideas, rather than trying to generate as many as possible. So, presumably, focussing on other people's ideas ("cognitive fixation") is a benefit.
    My first thought too. Having this ability to fixate and build on an idea presented by someone, is a great benefit to role-playing games.
    Posted By: GrahamYes, but that's not brainstorming. Brainstorming is, specifically, generating as many ideas as possible as a group. The evidence is that it's better to generate ideas individually, then bring them to a group.

    But that doesn't sound relevant to RPGs.
    Somehow it sounds relevant still ...
    - maybe the knowledge that this is how we create the most ideas, is something we may exploit in game-design? Working in liege with human nature seems like a good idea ...
  • Posted By: GrahamPosted By: stefoidI think the OP is referring to some games that have 'brainstorming' specfically as a part of play, such as -collectively outline an initial situationor something. Like in PTA , the first thing everybody does is agree on what type of TV series the game is going to be about.
    Yes, but that's not brainstorming. Brainstorming is, specifically, generating as many ideas as possible as a group. The evidence is that it's better to generate ideas individually, then bring them to a group.

    But that doesn't sound relevant to RPGs.

    For the purpose of this discussion, its better to think of brainstorming as a process rather than an aim. I think its a good point.
  • edited November 2011
    Any number of collaborative processes see this as a benefit, not a problem. World-building in Awen is blatantly this, and I assume (correct me if I'm wrong) that Microscope and Unversalis, in play, are beneficially and obviously driven by just this thing.

    (But, no, it's not brainstorming, in the classical sense. It's this other thing.)
  • Their hypothesis was right: When exposed to many cues, the undergrads offered up less creative, diverse ideas. The numbers improved when the students were given a five-minute break during the exercise.
    This is both good and bad. Good in that it can be used to focus mood and theme. Bad in that it can lead to less creative input. I've seen RPGs with lots of rules and concepts literally leave less 'room for imagination' during play. It's the difference between a character being able to poison someone by drilling through the floor and using a string to drip poison into a cup below and ONLY being able to poison someone if a character has the Use Poison skill.
  • So, people generate more diverse ideas on their own than in a group. That is indeed relevant to roleplaying, even if it turns out we don't often do brainstorming per se.

    Question: To what extent are diverse ideas helpful? At what point in a game are they helpful? Is it sometimes better to have less diversity?

    My answer would be that the typical rpg needs structured diversity. The various players around the table need to be on the same page. So in the type of situation where a "brainstorm" might be useful (generally around the point the game is set up), at the very least I'd say you want someone (the GM or the host or whoever) to give the group a remit for their brainstorm - limits within which to work. Then perhaps you could send everyone off to generate ideas on their own.

    Then again, isn't the reality that roleplayers, when doing the type of game that requires brainstorm-esque activity, are doing it precisely because they enjoy creating things as a group? Maybe it's worth sacrificing a bit of diversity of ideas for that.

    One more thing: This experiment simply shows that when you put people in a discussion with strangers there's inhibition of idea generation. The same effect might well not apply in a room where everyone trusts each other and where there's a lot of support for genuinely diverse creativity (my experience is that this is not the case in other environments where one is brainstorming, despite the fact that this is ostensibly what the brainstorm is meant to foster). So is this more relevant to con games, where people don't know each other and there's danger of social inhibition taking over? (And if so, is that a bad thing? Maybe you don't want your most wild and creative play to happen in a brand new group?)
  • Firstly, nice! Washington Post picked up on my report again - must be the third one this year.

    Secondly, it's a shame that it missed one of the nuances of the original academic article, that I tried to cover in my report - that the brainstorming groups went deeper into the categories that they did explore; where they suffered was in breadth of categories.

    I think this corresponds quite nicely to rpgs. As people are already discussing, we aren't necessarily interested in copious volumes of disparate ideas when eg worldbuilding. Rather, we want to get on the same page and generate some interesting possibilities from there.

    So given "Period:1930s", we can generate ideas alone and then come together: this is more likely to end up with ideas that include everything but the kitchen sink. Or, we meet fresh and generate ideas together, in which case we are more likely to hone in on certain themes or situations, and riff off of these, even without a firm hand (eg The Producer) encouarging this, simply because we are primed by the other early ideas to fixate on key areas.

    The risk is that those key areas end up being the obvious ones - as film noir often springs to mind first, we end up with the third noir game in as many months. But all that's needed to prevent that is some early ideas or framing that take us away from there.
  • Very cool Alex!
  • Posted By: GrahamBut that doesn't sound relevant to RPGs.
    If not for play, then I wonder if it is for design.

    - Ryan
  • I think perhaps this topic is relevant to parts of a game (1) which could be done as a group or individually, and (2) where diversity and uniqueness trumps consensus.

    For example, character creation:

    In games with strong pre-set settings and idioms, cohesion may be the mark of a good group of characters, and so, hey, make them out loud, together.

    But in a game that starts with "make a protagonist you're really excited to play" and only then moves on to establishing settings and situations for these characters, perhaps the characters should be made separately in silence.
  • Brainstorming is certainly relevant to play, especially in games such as Mortal Coil or PTA where this happens at the start of the game. MC has more support in the shape of questions which guide the discussion, PTA is a bit more of a free for all.
  • Well, RPGs doesn't stop at the end of Brainstorming. Usually it starts from there.
    There should be plenty of possibility to re-elaborate those ideas in a cycle that can bring all of that potential to the game.

    Anyway, I am starting to look at RPGs as a exclusive "group activity", and the "actor" of this activity is the Group, as a plural entity.
    For RPGs it is good that ideas and contribution stay within the limit of the Group capacity of expression.

    Building pieces of the game separately produces a set of... "soloists" that will struggle with a variety of degree of (un)success to find a spot for themselves within the cloud.
    Working already within the cloud is far way more productive and... well... "functional" (in all the way this may word may be used).
  • I think about this a lot when developing "brainstorming" (which it isn't, really, is it?) tools for my games.

    For example, in my game Musette, I have each person come up with one "story element" completely on their own. We get a series of things that are cool but may or may not go together. From there, I have the first player pick two elements (which are not their own) from the list that look they would go well together. If the group supports the pairing, it becomes "official".

    So I might look at our list of "1930's" elements, and pick out two that jump out at me, "recently unemployed banker" and "alien invasion". That narrows down our field immediately, and in a very interesting way: it suggests "average Joe" involved in some kind of weird story outside his control, maybe some "body snatchers", etc.

    There are lots of ways to approach this kind of constrained creative process, and I find them really interesting to explore.
Sign In or Register to comment.