The Psychology and Sociology of Creativity and IP, Stanford Law School

Session 1: Why do people create?

Discussion leaders: Jessica Silbey

Why and how? Psychological perspective, individual: b/c it’s fun, play; problem-solving; creativity as essential to who they are/identity-forming; as natural.

Literature on extrinsic v. intrinsic motivation. Different ways of operationalizing extrinsic motivation v. intrinsic.  Methods of studying: how are we observing motivation, how do we collect and interpret that data? Psychology and sociology have diverse ways of doing this.

Circumstances under which creativity happen: Amabile, Hyde on lack of constraints—less boundaries, more spatial freedom; Mihaly Cz. theory of “flow,” Amabile on constraintless play. 

Sociological perspective: doesn’t take the unit of measurement as the person but as the structure. Organizational form: the Linux corp.  Organizational processes: particular practices inside the organization, such as “release early and often.”  Technological processes: mechanisms w/in the rituals that support creativity—flows of paper or machines.  In the Linux study: the version control software and how that process changed the organization.  Final level: products—sociologists infrequently, but do, study outputs to understand how the organization works.  E.g., the chain store as an innovation of retail; product of a social structure of particular kinds of markets.

Laura Pedraza-Farina: Sociologists study how science happens. Communities of scientists have social norms, mostly pro-innovation; regulate how research gets done.  Don’t distinguish basic/applied; looking at what scientists actually do in the lab and in industry.

Sociology of expertise: another interesting field studying organization of expertise; sociology of networks also relevant.  Very interested in the study of boundaries: communities, unity of innovation. Given different names by different disciplines: social worlds, invisible college, communities of practice. So definition of community is important; articles are preoccupied with boundaries: how created, maintained; how does knowledge flow or not flow b/t communities?  Search for information, recombination of info is constrained by boundaries of community, and innovation comes from recombination.  Dan Burk on boundary objects: patent silences.  Objects can help breaches.

New research on sociology of networks. Two traditions. (1) To innovate you need cohesion and trust. (2) To innovate you need to be at a place where you have access to knowledge from elsewhere: you need to broker connections b/t communities.  New research on “structural folds”: you need both.  Overlapping membership in groups to get trust; social bonds to promote trust; but also tension from coming from different cognitive frames.

Silbey: concepts of what counts as creativity in these communities are also contested.

Commentators: Chris Buccafusco: Silent tertium quid here: economics. Psychology and sociology of IP=what a lot of us are thinking about is how these things relate to our assumed ideas about the economics of IP.  Psych and soc recast, reframe, challenge the economic story.  Neoclassical economics; new institutional economics will bring in the new concerns about groups, societies, infrastructure.  Also helps us think about what we mean by “psychology”—are these questions for behavioral/experimental econ?  Or “pure” psych?  How do we map these methodological tools for measuring why/what people create.

Breakdowns along subdisciplinary lines. Psych: cleavages b/t social and personality psychologists, who ask very different questions and have different answers. Personality psychologists will tell stories about the creative person and what makes people distinct.  Social = nature of institutions, nature of social relations.  As primarily legal scholars, we have choices.  How should we understand these different answers?  Which questions should we follow up on to give us purchase on questions for legal scholars/regulators?  Law tends to deal w/institutions; tends not to do much if it thinks that people are just very different from each other, if for example there are “creative people” and “others.”

What do people choose to create? Do they create stuff we care about, stuff that furthers IP’s goals?  Overproduction of creative goods and the relationship b/t overconfidence of individuals.  We hypothesize, that unlike mug endowment effect, people really like their paintings before they paint them (though Dan Burk suggests maybe not after).  How people think about their choices in the creative process.  How do risk profiles and risk influence creative production?  There are differences in personality w/r/t risk, but also differences in situation. Cognitive diversity in risk frames is really important. When creators think about designing around, litigation risks, market risks—we need to know more.

Chris Sprigman: What our scholarship is supposed to be for. W/in sociology and psych, there are people trying to understand what intrinsic motivations are and how people respond. Another strand: behavioral economics, trying to understand whether the neoclassical framework holds up.  Interaction so far: people in psych/soc have fed the people in law & econ doing behavioral work trying to deepen neoclassical account. We haven’t been feeding them; how can we be better at talking to our colleagues and giving them reasons to be consumers of our work?

Neoclassical story is attractive b/c it’s linear and gives you answers/policy suggestions.  Institutional econ complexifies the story.  Payoff of virtually every paper that Sprigman & Buccafusco have worked on is: it’s more complex than that, to which the answer has been “ok, great.”

Could say: the aim is to describe the world, and it is what it is. But policymaking is more complex than that. Eckartz et al., how do incentives affect creativity. Creativity game for making words out of random letters. Flat fee v. pay per performance v. tournament—if you do very well, big payment, but otherwise nothing.  Tournament type games are very invigorating, hypothesis goes—but it turns out that’s not true; people tend to do about the same. People tend to self-sort into flat fee.  Performance is driven by individual differences not payment scheme. They perform well b/c they’re good at doing the task. People better at the task tend to self-sort into pay for performance groups, but they don’t do any better in those groups.

So, under what conditions would a business be well advised to pay a flat fee?  This requires more investigation.

Buccafusco/Bechtold/Sprigman paper: creativity/innovation heuristics: we make people play different types of creativity games—we look at decision to innvote v. borrow, though it’s really borrowing more/borrowing less. See if people respond to real money incentives to do more of one or the other. The answer: it doesn’t seem to do much to move people’s behavior. What’s the determinant?  In Eckharz it seems to be self-perception of competence at game. In our study, it seems to be whether they perceive the problem to be hard.  If you can control decision on innovating/borrowing by varying subjective perception of the problem, how do you frame the problem to do that?  Framing effects may have much more to say than incentives, b/c even when we made incentives supercompensatory it moved innovation a little not a lot.

Neoclassical economics doesn’t hold up well.  It can fight back—say you’re just detecting a richer preference set—so it will be an ongoing discussion. More to the point, what positive contribution that we can make as our distinctive contribution—connection to policy?  People don’t like complexity!

Julie Cohen: We may still have difficulty putting our finger on the cultural—it’s in the water (fish says: what the hell is water?); we’re in a mad rush to colonize it with numbers and empirical studies, but culture is the everything else that surrounds us. The way to get access to it is not with running studies but by reading literatures in big traditions in philosophy, cultural studies, media studies, and allowing ourselves to recognize and admit into the room the less deterministic and less easy to measure things like archetypes, memes, values, linguistic traditions.  Remember in our systematization that there is also such a thing as culture.

Deterministic ways of thinking about these processes and and seeming congeniality of determinate/deterministic methods: in my work, what’s as important as structure is the absence of structure. Play as serendipity/freedom to combine disparate things into structures.  When I read work that says constraint is important so we should load on constraints, it makes me unhappy (me too!) because we need to acknowledge the role of degrees of freedom/play.

Endogeneity of all these factors.  Merges & Nelson on patent scope, Burk & Lemley on policy levers—if you read these side by side, you see M&N say patents are unsuited to software b/c corporate capital isn’t involved and it’s individual; B&L 20 years later say that patents are perfectly suited—b/c institutions have changed, everything else has changed in terms of economic analysis, policy analysis.  Another example in ©: everything is trolls now. ASCAP was a troll.  Then today it’s at the table every time Congress does anything and no one would dare call it a troll.  [Maybe Ann Bartow would.]  All the variables are endogenous. Change institutional structures and people are still creating, but differently; software has become domesticated/disciplined w/in the patent framework and the industry looks different.  Think about alternative paths; we tend to take current institutions as inevitable and it’s worth remembering that everything could have been different and creativity would not have ended but would have been different; have some humility.

Silbey: Bibliography: there’s a huge literature on risk from the insurance perspective and sociology of professions: different professions diagnose and manage risk very differently.  Stereotype threat: if you tell girls they’re bad at math, they do worse on test.  That’s actually related to Buccafusco/Sprigman work.  That’s important for understanding how we situate ourselves in our universities. 

Jeanne Fromer: The literature is complicated; there is no one-sentence summary about creation and incentives. It’s unfortunate in some ways that we’ve decided to conclude that incentives diminish creativity; people have picked up on that thread, which exists and is prominent in Amabile’s work, but it’s much more nuanced, including in her earlier work.  Chris Eisenberger’s meta-study: when you give people incentives and a goal, incentives help; if you don’t give them a goal, they were less creative w/incentives.  I’m probably decontextualizing it more than necessary. We’ve run to legal conclusions faster than we should. Dialogue w/many in the field, b/c there is diversity, is something we need to do carefully.

Reason to have the dialogue: it’s important for them to understand legal institutions. To the extent they’re interested in this world, they often ask their questions in ways that are a bit askew/not fully legally salient. So we need to work with them.

David Fagundes: All the experiments are about individual response to incentives. One twist: vast majority of decisions to create in high $ areas are made by companies.  Is there any reason to think that institutions come out differently on the incentive front than individuals? Maybe institutions respond better to extrinsic motivations.  [Though institutions are made of people.]  We insist on having a unitary theory of incentives, and that doesn’t need to be the case. Maybe if there is something foundational about the individual/institution divide, we need different incentives; maybe the current system incentivizes institutions as we want, and everybody else you can’t do much about so we don’t need to worry about them.

Mark Lemley: One level at which you could read this literature is to say: the evidence about money as motivation is not great, but corporations are motivated by money so it all works out. The literature on commercialization explanations for IP is even weaker than the personal intrinsic motivations explanation.  One thing in particular: need for IP in many industries is going away as distribution costs go to zero. There’s a variant that it’s not really commercialization—it’s that “we hire these creative people only b/c we make money,” but that leads to the Q whether these people create better w/in institutions that have hired them to create.

Complexity: what’s the psychology of the response to “it’s complicated.”  How does that get filtered into policy?

Rob MacCoun: If you ask me to perform in celestial mechanics, I’ll perform as well for cheese sandwich as for $1 million; individual differences matter. In equilibrium, a firm will find people who have ability, but experiments take as a given people who are there.  Work on people underestimating their own ability: people leave money on the table b/c they don’t realize they’re more efficacious.

Jim Bessen: Power loom improved productivity; many improvements were unpatented, unpatentable, tacit skills, that increased productivity even more astonishingly. Most of the creativity wasn’t even the inventor’s.

Greg Mandel: Complexity is a difficult sell in policy, but most fields of law involve highly contested justifications for their bases.  Take criminal law: deterrence, retribution—we still try to study it as deeply as we can before we get to policy.  Disputes in the field around whether creativity is domain-specific or not; there’s a strong school of thought that there’s holistic similarity b/t artistic and scientific creativity, but there are domain-general types of creativity—empathy and communication v. science and tech v. art.  W/in those fields, there are both domain-general and domain-specific kinds of creativity.

Abishek Nagaraj: Economics differs from psych/soc in terms of welfare. Most economists aren’t interested in innovation per se, but in the implications.  Connect differences in innovation to other outcomes, such as a price impact. This goes to “what counts” as innovation—the onus is on the researcher to point out welfare implications of differences in creativity.  Understand completely that it will be context-specific; one approach would be to directly try natural/archival data; if you study a large enough data set, that is useful.  Flat v. pay for performance: you might get complicated answers in the lab; there is work on HHMI grants (flat) versus NIH (more short term). He finds flat incentives lead to more creative work; even if those answers don’t apply to mugs/individuals, they’re useful in themselves.

Mark McKenna: Why do we care what other literature has to say?  We are interested in how people actually behave.  Or maybe the stories we tell in law are actually normative, not empirical. One reason to engage is holding a mirror up to law & econ to expose its normativity. IP law isn’t remotely unique in that way.  We’ve learned similar things about how teachers respond to teaching bonuses (they hate them).  There’s no such thing as “more” creativity. There’s more of some things and less of others, and we’re making choices; one reason to engage is that we like to have stories about how our system is one size fits all. We should be more explicit about what we want and why.

Madhavi Sunder: Why is creativity a social good that we think it’s important to regulate the production and distribution of in order to promote social welfare?  Other disciplines can help us structure—incentives for the individual are the tree, but we should look at the health of the whole forest.  Effects on other creators.

Laura Pedraza-Farina: Sociology has specific definitions of innovation. Do we agree with those metrics? 

Andrew Torrance: Schumpeter talks about psych, sociology, and economics, w/only a law degree; he was able to embrace all. It would be useful to distinguish some of the things Cohen has talked about—making phase v. diffusion phase, with different implications from psych/sociological perspectives.  Ruth Schtalk (sp?) looked at “big five” personality traits and conception, prototyping, and diffusion phase in survey of German innovators. Openness and risk-taking come up with great conceptions. Prototyping: they drop out and introverts/high conscientiousness matters.  Diffusion: P2P diffusers have low conscientiousness; traditional IP users have high conscientiousness. Innovation can be broken apart, and psych has a lot to say about why we know so little about where to apply the legal tools in various phases.

Mark Suchman: Policy pushes us to less complex accounts b/c they seem to cover everything.  Law is a shaper of attention.  HIPAA has created a huge level of response in the industry far beyond rational response to likely enforcement. [Institutional accounts: the place of the privacy officer!] Law provides categorical responses that then structure behavior.  Software: © v. patent is a good example.  These are defaults that affect behavior not through incentives provided by law but through categories provided by law.

RT: Julie’s right as usual, and I want to emphasize that there is a gendered aspect to how we talk about creativity, or really a series of gendered aspects.  Take Dave Fagundes’ point: Institutions respond to incentives: well, sort of.  Like Soylent Green, instutions are made of people, and they often respond only to incentives that aren’t threatening to the individuals who are in charge of the institution at that time.  (Compare the response of the music industry to the rise of digital.)  Dollar for dollar return on movies w/female leads: talk about leaving money on the table!  J.K. Rowling, Stephenie Meyer, E.L. James: there is a market, but institutions didn’t really believe in it for a long time.  That’s not just Schumpeterian creative destruction; there’s a reason that these are all women, and that similar stories exist with rap. Teenage girls are driving linguistic changes as well as artistic changes, but there’s nothing that many grownups like to disparage more than teenage girls. 

The “we don’t need to worry about creativity outside the market” position is another version of “housework and childcare are naturally provisioned for free so we don’t need to think about them in our policies.”  Look at literature on discrimination: why didn’t the free market get rid of the gender and race wage gaps?  That’s actually a similar question to why we see differences in creative output that are associated with demographic categories of interest.  (See Robert Brauneis & Dotan Oliar’s recent work on copyright, race, gender and age.)  Cultural theory, feminist and critical race theory.

Ben Depoorter: Does the experiment support a broad, sweeping statement the way we see in the literature? Tension b/t economists and behavioral economics. Demand story: many of us want to publish in law reviews, which requires us to generalize/have ambitious and broad policy recommendations. We can change our publication priorities—subject ourselves to peer review by psychologists as a necessary step.

Stephanie Bair: Extrinsic v. intrinsic—there is a type of motivation that’s technically extrinsic but is a great motivation, which is accolades from engaging in creative work.  Lots of people have landed on attribution/reputation as a good thing for IP systems.  Other literature: Think a task is easy = many people have overconfidence bias; hard = underconfidence bias. Supports the Sprigman/Buccafusco results.

Silbey: case studies are great idea, as Nagaraj said.

Tim Holbrook: How do these incentives, assuming they exist, get translated through the institutional/corporate structure? Sociological and linguistic question.  Rules of thumb get translated—Emory had advice about “how many words” you could use in fair use.  Telephone game has gotten more complex over time.  Does this translation from general to specific account for some of this complexity, allowing institution-specific tailoring?  That said, the way that this translation happens clearly has social/racial/gender dynamics, such as how people undervalue themselves in gendered ways.

Fagundes: Maybe we can’t be as confident that new institutions respond in the same way as old ones.  But there is still a money-seeking attitude: we will make crap as long as you buy it.  I remain skeptical that there’s a unitary motivation in what drives people to create.  One distinction I haven’t seen teased out is what motivates people to initially invest v. what continues to motivate them to invest throughout. These aren’t the same, esp. for institutions.

Lemley: Case studies are great, but not if they are used by policymakers to set the rules for everybody—they’re often context-specific. 

Kate Darling: Legal scholarship isn’t as good as other disciplines at the bigger picture; we’re able to find issues that are relevant to policy, and to evaluate what evidence will actually influence policy. 

Buccafusco: Descriptive and normative work are both important and we often do both, but we must still distinguish them.

McKenna: Institutions leaving money on the table for gendered reasons: that’s clearly a response to incentives, they just aren’t monetary incentives.  This is a richer set of preferences that we tend not to integrate into our economic accounts. We have a hammer and want to look like nails; our legal tools are less well positioned to deal with those incentives.

Paul Goldstein: Underlying question of framing the Q of creativity.  Can frame it generally: across technology and literary/artistic creativity.  Suspicious of studies about incentives for individual creators.  Spending time w/working novelists: Ego makes them work, whether they could make a living with their writing alone or not.  We tend to look at creativity/creative production as episodic rather than as a continuum; novelists’ reaction to the first draft though is often horror.  [Here is my favorite point about that, though I’m also a firm believer in fixing the bad first draft.]  What does it take to actually get from draft 1 to 25?  It’s not necessarily economic incentives but ego.  Leave the individual creator alone to exercise her ego; the answers lie in ©’s ability to control initial publication, and in human rights.

Julie Cohen: Asking for determinate policy provisions is only one question one could ask in this space. It’s clear in light of cultural/institutional evolution that lots of different configurations generate creativity. “What does this description tell us we should not do?” is an equally important question.  Many policy propositions may be feasible, but some perhaps should be put off limits.

Stephanie Bair: one response to “individuals don’t need money incentives” is “that just screws individuals v. big companies again.” A policy Q: how do we make sure innovative and creative people are able to make a living?  [Well, we can’t do that with IP policy.  You can build it; they don’t have to come.]  If we paid everyone to follow their passions, who would sweep the streets? [Basic income? Or, as Sprigman said, raise the wages for street sweeping?]

Suchman: There are © systems (authors’ rights systems) that give more weight to authors’ ego.  We think about academic scholarship, but notice numbers of authors on average law review, sociology, physics article—obviously when you’re the 300^th author on a physics paper, it’s important to your career and ego but perhaps in a different way.  Questions about who has space to be creative, resources—as much about organizational politics as anything else—are people on the shop floor allowed to make productivity enhancing changes, rewarded for it, fired for it, given credit for it?

Lemley: Lurking issue: why do we want creativity?  Not obvious we want people to make a living doing creative things. Huge numbers of people create good stuff not b/c they’re paid to do it, and not b/c they make a living from it, w/other motivations. Maybe we want a pro class paid to create b/c we get different creativity and different is good, or b/c we think pro creativity is better, but this is an empirical question.  Is the pro class actually what we want as a step towards some instrumental goal of more/better creativity?